Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a data sharing method, system, device and medium based on a virtual power plant to solve the above-mentioned problems in the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
A data sharing method based on a virtual power plant comprises the following steps:
s1, acquiring real-time operation data of distributed energy resources of a virtual power plant through a shared communication network, wherein the real-time operation data comprise resource type identification and adjustment cost parameters;
S2, judging the transmission delay state of real-time operation data in the shared communication network, and generating a data stream set of delay marks and non-delay marks;
S3, for non-delay marked data, matching a preset initial priority rule based on a resource type identifier, for the delay marked data, determining an attenuation factor according to the resource type identifier and the grid frequency modulation demand level in a cooperative manner, and generating a dynamic priority compensation coefficient based on an adjustment cost parameter and the attenuation factor;
s4, fusing the initial priority rule and the dynamic priority compensation coefficient to generate a real-time dynamic priority sequence;
S5, simulating the influence of different dispatching instruction combinations on the physical safety constraint index of the power grid based on the real-time dynamic priority sequence, and generating a safety margin assessment result;
And S6, adjusting the real-time dynamic priority sequence according to the safety margin evaluation result, generating a resource scheduling instruction and distributing the resource scheduling instruction to corresponding resource execution through the shared communication network.
In a preferred embodiment, acquiring real-time operational data of a virtual power plant distributed energy resource over a shared communication network, including resource type identification and regulatory cost parameters, includes:
Acquiring real-time operation data of distributed energy resources of the virtual power plant at a preset acquisition frequency through a shared communication network;
converting the real-time operation data into a unified data format, wherein the unified data format comprises a resource type identification field and an adjustment cost parameter field;
and carrying out validity verification on the resource type identification field and the adjustment cost parameter field, and eliminating invalid data entries.
In a preferred embodiment, determining a transmission delay status of real-time operational data in a shared communication network, generating a set of delay marked and non-delay marked data streams, includes:
acquiring a time stamp of real-time operation data, wherein the time stamp comprises data generation time and receiving time;
Calculating a transmission time according to the data generation time and the receiving time, wherein the transmission time is equal to the receiving time minus the data generation time;
comparing the transmission time with a preset delay threshold, marking the data with the transmission time exceeding the preset delay threshold as delay marking data, otherwise marking the data as non-delay marking data;
And classifying the delay marked data and the non-delay marked data according to the resource type identification, and generating a delay marked data stream set and a non-delay marked data stream set classified according to the resource type.
In a preferred embodiment, for non-delay tag data, matching a preset initial priority rule based on a resource type identification includes:
acquiring a preset initial priority rule table, wherein the initial priority rule table comprises a mapping relation between a resource type identifier and an initial priority value;
Inquiring an initial priority rule table according to a resource type identification field in the non-delay marked data, and matching a corresponding initial priority value;
weighting and summing the matched initial priority value and the adjustment cost parameter in the non-delay marked data to generate an initial priority score;
Writing the initial priority score into a priority field of the non-delay marked data to generate a non-delay data stream set with the initial priority score;
generating dynamic priority compensation coefficients for delay marker data, comprising:
determining a reference attenuation factor according to the frequency modulation demand level of the power grid, wherein the frequency modulation demand level of the power grid is an emergency reference attenuation factor which is a first preset value, a second preset value in normal conditions and a third preset value in loose conditions;
Dynamically adjusting the reference attenuation factor based on the resource type identifier corresponding to the delay mark data and the real-time schedulable capacity, if the resource type identifier is energy storage and the real-time schedulable capacity is larger than a capacity threshold, increasing the compensation amount by the reference attenuation factor, otherwise, reducing the compensation amount;
Taking the ratio of the dynamically adjusted reference attenuation factor to the adjustment cost parameter as a dynamic priority compensation coefficient;
and carrying out secondary correction on the dynamic priority compensation coefficient according to the power grid node voltage out-of-limit risk level, and if the power grid node voltage out-of-limit risk level is high risk, applying an inhibition factor to the dynamic priority compensation coefficient.
In a preferred embodiment, fusing the initial priority rule with the dynamic priority compensation coefficient generates a real-time dynamic priority sequence comprising:
Classifying and layering the initial priority score of the non-delay data flow set and the dynamic priority compensation coefficient of the delay data flow set according to the coupling relation between the frequency modulation demand level of the power grid and the resource type identifier;
The layering priority order is dynamically adjusted based on the real-time power grid frequency deviation change direction, wherein energy storage type resources are preferentially called when the frequency is positively deviated, and load type resources are preferentially called when the frequency is negatively deviated;
Carrying out multidimensional cross sequencing on non-delay data and delay data according to the resource type identification and the grid frequency modulation demand level, and generating a priority subsequence of the resource type-frequency modulation level dimension;
The priority subsequence is safely screened according to the power grid node voltage out-of-limit risk level, namely resource items which are likely to exacerbate the voltage problem are removed under the high risk level, and the complete priority subsequence is reserved under the low risk level;
And merging all the priority subsequences after the safety screening, and generating a real-time dynamic priority sequence according to a global priority rule of the grid frequency modulation demand level and the resource type identifier.
In a preferred embodiment, simulating the influence of different scheduling instruction combinations on the grid physical safety constraint index based on the real-time dynamic priority sequence, and generating a safety margin evaluation result comprises:
Generating a plurality of candidate scheduling instruction combinations according to the real-time dynamic priority sequence, wherein each candidate scheduling instruction combination comprises scheduling instructions with different resource type identifiers and corresponding adjustment quantity parameters;
Performing power grid physical security constraint simulation on each candidate scheduling instruction combination based on current operation parameters of the power grid, and predicting power grid physical security constraint index change values after the candidate scheduling instruction combination is executed;
comparing the predicted power grid physical security constraint index change value with a preset power grid physical security constraint index security threshold value to generate a security margin evaluation value of each candidate scheduling instruction combination;
and screening candidate scheduling instruction combinations with the safety margin evaluation value larger than or equal to a preset safety margin evaluation threshold value, and generating a safety margin evaluation result set.
In a preferred embodiment, the method for adjusting the real-time dynamic priority sequence according to the safety margin evaluation result, generating a resource scheduling instruction and distributing the resource scheduling instruction to corresponding resource execution through a shared communication network comprises the following steps:
Adjusting the real-time dynamic priority sequence according to the safety margin evaluation value in the safety margin evaluation result set, and if the safety margin evaluation value of the candidate scheduling instruction combination is lower than a preset safety margin evaluation threshold value, eliminating the corresponding resource item from the real-time dynamic priority sequence;
Generating a resource scheduling instruction based on the adjusted real-time dynamic priority sequence, wherein the resource scheduling instruction comprises a resource type identifier, an adjustment quantity parameter and an execution time window;
Distributing the resource scheduling instruction to corresponding distributed energy resource terminal equipment through a shared communication network;
After the distributed energy resource terminal equipment receives the resource scheduling instruction, executing the adjustment quantity parameter according to the control logic corresponding to the resource type identifier, and feeding back an execution result to the virtual power plant aggregation control platform through the shared communication network.
In another aspect, the present invention provides a data sharing system based on a virtual power plant, including:
The real-time acquisition module is used for acquiring real-time operation data of the distributed energy resources of the virtual power plant through a shared communication network, wherein the real-time operation data comprise resource type identification and adjustment cost parameters;
The delay marking module is used for judging the transmission delay state of the real-time operation data in the shared communication network and generating a data stream set of a delay mark and a non-delay mark;
The dynamic compensation module is used for matching a preset initial priority rule based on the resource type identification aiming at the non-delay marked data, determining an attenuation factor according to the resource type identification and the grid frequency modulation demand level in a cooperative manner, and generating a dynamic priority compensation coefficient based on the adjustment cost parameter and the attenuation factor aiming at the delay marked data;
the sequence generation module is used for generating a real-time dynamic priority sequence by fusing the initial priority rule and the dynamic priority compensation coefficient;
The safety evaluation module simulates the influence of different dispatching instruction combinations on the physical safety constraint index of the power grid based on the real-time dynamic priority sequence, and generates a safety margin evaluation result;
And the instruction distribution module is used for adjusting the real-time dynamic priority sequence according to the safety margin evaluation result, generating a resource scheduling instruction and distributing the resource scheduling instruction to corresponding resource execution through the shared communication network.
On the other hand, the invention provides data sharing equipment based on the virtual power plant, which comprises a processor, a memory and a program or an instruction stored on the memory and capable of running on the processor, wherein the program or the instruction is executed by the processor to realize the data sharing method based on the virtual power plant.
On the other hand, the invention provides a data sharing medium based on the virtual power plant, wherein a program or instructions are stored on the medium, and when the program or instructions are executed by a processor, the data sharing method based on the virtual power plant is realized.
Compared with the prior art, the invention has the following beneficial effects:
1. The problem of multi-source cooperative failure caused by data asynchronism under the traditional static priority rule is solved by optimizing a dynamic sharing mechanism of the multi-source data stream, the system dynamically identifies delayed and non-delayed data streams based on the acquisition and transmission state marks of the shared communication network on the real-time operation data, and generates a priority compensation coefficient by adjusting cost parameters and frequency modulation demand level, so that delayed data of low-cost resources can still participate in scheduling decisions through dynamic weight adjustment, cooperative optimization of multi-source heterogeneous data under a transmission delay scene is realized, and resource error scheduling risk caused by data stream asynchronization is reduced;
2. The closed-loop linkage of the data flow state mark and the power grid safety constraint further strengthens the guarantee capability of the data sharing process on the power grid operation safety, and after the priority sequence is dynamically generated, the data sharing weight is reversely corrected based on the safety margin evaluation, so that the dispatching instruction is ensured to meet the economic target, the physical safety boundary of the power grid is met, the autonomous coordination capability of the virtual power plant on the multi-source data in the high-frequency transaction scene is improved, and the reliability of the complex power market environment is improved.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment 1 fig. 1 shows a data sharing method based on a virtual power plant, which comprises the following steps:
s1, acquiring real-time operation data of distributed energy resources of a virtual power plant through a shared communication network, wherein the real-time operation data comprise resource type identification and adjustment cost parameters;
S2, judging the transmission delay state of real-time operation data in the shared communication network, and generating a data stream set of delay marks and non-delay marks;
S3, for non-delay marked data, matching a preset initial priority rule based on a resource type identifier, for the delay marked data, determining an attenuation factor according to the resource type identifier and the grid frequency modulation demand level in a cooperative manner, and generating a dynamic priority compensation coefficient based on an adjustment cost parameter and the attenuation factor;
s4, fusing the initial priority rule and the dynamic priority compensation coefficient to generate a real-time dynamic priority sequence;
S5, simulating the influence of different dispatching instruction combinations on the physical safety constraint index of the power grid based on the real-time dynamic priority sequence, and generating a safety margin assessment result;
And S6, adjusting the real-time dynamic priority sequence according to the safety margin evaluation result, generating a resource scheduling instruction and distributing the resource scheduling instruction to corresponding resource execution through the shared communication network.
Acquiring real-time operation data of distributed energy resources of a virtual power plant through a shared communication network, wherein the real-time operation data comprise resource type identification and adjustment cost parameters and comprise the following steps:
Acquiring real-time operation data of distributed energy resources of the virtual power plant at a preset acquisition frequency through a shared communication network;
converting the real-time operation data into a unified data format, wherein the unified data format comprises a resource type identification field and an adjustment cost parameter field;
and carrying out validity verification on the resource type identification field and the adjustment cost parameter field, and eliminating invalid data entries.
The preset collection frequency is set according to the scheduling requirements of the virtual power plant, including but not limited to once per second for the energy storage device and once per minute for the controllable load. In specific implementation, the collection frequency is configured through a virtual power plant aggregation control platform, and the aggregation control platform dynamically adjusts the collection frequency according to the resource type identification, for example, when the illumination intensity fluctuation of the photovoltaic equipment exceeds a preset threshold value, the collection frequency is automatically lifted to twice per second.
The unified data format adopts a JSON structured data format, wherein a resource type identification field is named as 'resource_type', a field value is named as a preset enumeration type (comprising 'stored energy', 'photovoltaic', 'controllable load'), an adjustment cost parameter field is named as 'cost_parameter', a field value is named as a floating point number, and the adjustment cost parameter field is used for representing the cost of unit adjustment quantity (unit is meta/kWh). In the implementation, the data format conversion is realized through an edge computing node, after the edge computing node receives the original data, the edge computing node analyzes the equipment number in the original data according to the resource type identification, queries a pre-stored adjustment cost parameter mapping table, and maps the equipment number into a corresponding adjustment cost parameter value.
Validity verification includes rules that the value of the resource type identification field must be within a preset enumeration type range, and that the value of the adjustment cost parameter field must be greater than or equal to zero and less than a preset upper threshold (e.g., 100 yuan/kWh). In specific implementation, the edge computing node executes verification logic on each piece of data, if the resource type identification field is 'energy storage' but the adjustment cost parameter is greater than 50 yuan/kWh, the data are judged to be abnormal data and are rejected, and if the resource type identification field is an undefined enumeration value (such as a fan), the data are marked to be invalid data and are discarded. After invalid data is removed, the remaining data items are transmitted to a virtual power plant aggregation control platform through a shared communication network for processing in the subsequent steps.
Notably, the shared communication network can be a multi-master collaborative 5G/blockchain network.
Judging the transmission delay state of real-time operation data in a shared communication network, generating a data flow set of delay marks and non-delay marks, comprising:
acquiring a time stamp of real-time operation data, wherein the time stamp comprises data generation time and receiving time;
Calculating a transmission time according to the data generation time and the receiving time, wherein the transmission time is equal to the receiving time minus the data generation time;
comparing the transmission time with a preset delay threshold, marking the data with the transmission time exceeding the preset delay threshold as delay marking data, otherwise marking the data as non-delay marking data;
And classifying the delay marked data and the non-delay marked data according to the resource type identification, and generating a delay marked data stream set and a non-delay marked data stream set classified according to the resource type.
The data generation time is recorded when the terminal equipment of the distributed energy resource generates real-time operation data, a real-time clock chip is arranged in the terminal equipment, a time signal of the real-time clock chip is synchronous with a reference clock of the virtual power plant aggregation control platform through a network time protocol, and the time synchronization error is less than 1 millisecond. The receiving time is recorded by the virtual power plant aggregation control platform when the data is received through the shared communication network, and the recording precision is in the millisecond level. The format of the time stamp adopts the coordination world time, the data generation time is written into the generation_time field of the data packet head, and the receiving time is written into the recovery_time field of the data packet tail. For example, the terminal equipment of the photovoltaic equipment generates real-time operation data when the illumination intensity changes by more than 5%, and writes the generation time of 2023-10-05T08:00:00.500Z into the packet header, and the aggregation control platform records the receiving time of 2023-10-05T08:00:01.200Z after receiving.
When the transmission time is calculated and embodied, the virtual power plant aggregation control platform analyzes the generation_time field of the data packet header and the recovery_time field of the data packet tail, converts the two time stamps into millisecond time stamps of the same time zone, and performs subtraction operation to obtain the transmission time. For example, the energy storage device may be generated at a time of "2023-10-05T08:00:00.500Z", received at a time of "2023-10-05T08:00:01.200Z", and transmitted at a time of 700 milliseconds. If the time zone of the data generation time is inconsistent with the time zone of the receiving time, the aggregation control platform firstly converts the two time stamps into coordinated universal time and then calculates a difference value.
The preset delay threshold is set according to the resource type identifier, wherein the energy storage type resource is 500 milliseconds, the photovoltaic type resource is 800 milliseconds, and the controllable load type resource is 2000 milliseconds. The threshold setting is based on the maximum allowable delay of different types of resources, and the aggregation control platform pre-stores a threshold mapping table, wherein the table structure comprises two columns of resource type identification and preset delay threshold. For example, querying the mapping table when the resource type is identified as "stored energy" results in a threshold of 500 milliseconds. When the transmission time of a certain piece of energy storage data is 700 milliseconds, the aggregation control platform judges that the transmission time exceeds a threshold value and marks the data as delay marked data, and when the transmission time of a certain piece of controllable load data is 1500 milliseconds, the data is marked as non-delay marked data.
When the delay marked data stream set and the non-delay marked data stream set classified according to the resource types are generated and are implemented in a concrete way, the aggregation control platform extracts a 'resource_type' field value in the real-time operation data, and the delay marked data and the non-delay marked data are respectively stored in independent sets named by the resource type identifiers. For example, delay tag data, the resource type of which is identified as "store" is stored in a "store_delay" set, and non-delay tag data is stored in a "store_non-delay" set. Each set is stored in JSON array format, and the array elements comprise an original data packet header, a packet tail and a tag status field, and the tag status field value is "delayed" or "non-delayed". The upper limit of the storage capacity of the collection is set according to the resource type, the upper limit of the collection of the energy storage class is 1000, the photovoltaic class is 2000, and the old data is covered according to the first-in first-out rule when the upper limit is reached.
For non-delay marked data, matching a preset initial priority rule based on a resource type identification, including:
acquiring a preset initial priority rule table, wherein the initial priority rule table comprises a mapping relation between a resource type identifier and an initial priority value;
Inquiring an initial priority rule table according to a resource type identification field in the non-delay marked data, and matching a corresponding initial priority value;
weighting and summing the matched initial priority value and the adjustment cost parameter in the non-delay marked data to generate an initial priority score;
writing the initial priority score into a priority field of the non-delay marked data to generate a non-delay data stream set with the initial priority score.
The initial priority rule table is preset by the virtual power plant operator according to the historical scheduling efficiency of the resource types and the power grid regulation and control requirements, and is stored in a configuration database of the virtual power plant aggregation control platform. The table structure of the initial priority rule table comprises two columns of a resource type identifier and an initial priority value, wherein the resource type identifier is an enumeration value (such as an energy storage value, a photovoltaic value and a controllable load), the initial priority value is an integer from 1 to 10, and the higher the value is, the higher the priority is. For example, the initial priority value corresponding to the resource type is 8 when the resource type is identified as "energy storage", and the initial priority value corresponding to the resource type is 5 when the resource type is identified as "controllable load". The initial priority rule table is imported during the virtual power plant initialization phase and supports dynamic updating according to the grid operating state, for example, when the grid frequency deviation continuously exceeds 0.2Hz, the initial priority value of "energy storage" is raised to 9.
When the matching of the corresponding initial priority value is implemented, the virtual power plant aggregation control platform extracts a field value (such as 'energy storage') of 'resource_type' in the non-delay marked data, queries the initial priority rule table by taking the field value as an index, and returns the corresponding initial priority value (such as 8). If the resource type identification field value is not defined in the rule table (e.g., "fan"), the initial priority value of the data is set to a default value of 1. For example, a resource type of a certain piece of photovoltaic data is identified as 'photovoltaic', an initial priority value 6 is obtained by inquiring a rule table, and another piece of undefined resource type data is marked as a default priority value 1 and an alarm log is triggered.
When the initial priority score is generated, the weight of the weighted summation is dynamically set according to the resource type identifier, the weight of the initial priority value of the energy storage type resource is 0.7, the weight of the cost parameter is regulated to be 0.3, the weight of the initial priority value of the controllable load type resource is 0.4, and the weight of the cost parameter is regulated to be 0.6. In specific implementation, the virtual power plant aggregation control platform queries a preset weight mapping table according to the resource type identifier, for example, when the resource type identifier is "energy storage", the initial priority score=the initial priority value×0.7+ and the adjustment cost parameter×0.3. If the initial priority value of a certain piece of energy storage data is 8 and the adjustment cost parameter is 2 yuan/kWh, the initial priority score is 8×0.7+2×0.3=6.2. The weight mapping table is stored in the configuration database and supports dynamic adjustment according to the market price fluctuation rate, for example, when the real-time price fluctuation rate exceeds 10%, the weight of the adjustment cost parameter of the controllable load is increased to 0.8.
Writing the initial priority score into a priority field of the non-delay marked data to generate a non-delay data stream set with the initial priority score. The 'priority_score' field is added in the JSON format of the non-delay marked data, and the field value is the calculated initial priority grade. For example, the "priority_score" field of a certain piece of stored data is written in 6.2, and the photovoltaic data is written in 4.8. All non-delayed data with priority scores are stored in separate sets of data flows by resource type classification, e.g. "energy storage_non-delayed_scored" set and "controllable load_non-delayed_scored" set. The storage format of the data stream set is consistent with the delay marked data stream set generated in step S2.
For delay marking data, determining an attenuation factor according to the resource type identification and the grid frequency modulation demand level in a cooperative manner, and generating a dynamic priority compensation coefficient based on the adjustment cost parameter and the attenuation factor, wherein the method comprises the following steps:
determining a reference attenuation factor according to the frequency modulation demand level of the power grid, wherein the frequency modulation demand level of the power grid is an emergency reference attenuation factor which is a first preset value, a second preset value in normal conditions and a third preset value in loose conditions;
Dynamically adjusting the reference attenuation factor based on the resource type identifier corresponding to the delay mark data and the real-time schedulable capacity, if the resource type identifier is energy storage and the real-time schedulable capacity is larger than a capacity threshold, increasing the compensation amount by the reference attenuation factor, otherwise, reducing the compensation amount;
Taking the ratio of the dynamically adjusted reference attenuation factor to the adjustment cost parameter as a dynamic priority compensation coefficient;
and carrying out secondary correction on the dynamic priority compensation coefficient according to the power grid node voltage out-of-limit risk level, and if the power grid node voltage out-of-limit risk level is high risk, applying an inhibition factor to the dynamic priority compensation coefficient.
And determining a reference attenuation factor according to the grid frequency modulation demand level, wherein the reference attenuation factor is determined to be an emergency level when the absolute value of the real-time grid frequency deviation exceeds 0.5Hz, the reference attenuation factor is set to be 0.9, the reference attenuation factor is set to be a normal level when the deviation is between 0.2Hz and 0.5Hz, the reference attenuation factor is set to be 0.6, and the reference attenuation factor is set to be a loose level when the deviation is less than 0.2 Hz. The numerical value is determined by analyzing the adjustment success rate and the cost-benefit ratio of the energy storage resources in different scenes in the virtual power plant history scheduling database. For example, the average adjustment success rate of the energy storage resources in 100 emergency scenes (deviation >0.5 Hz) in the past year is counted to be 92% (i.e. more than 90%), corresponding to a reference attenuation factor of 0.9, the adjustment success rate of the normal scenes (deviation 0.2Hz-0.5 Hz) is counted to be 63%, corresponding to 0.6, and the adjustment success rate of the loose scenes (deviation <0.2 Hz) is counted to be 28%, corresponding to 0.3. The setting of the value ensures the resource priority in the high-success-rate scene to be improved and the low-success-rate scene to be suppressed.
For example, if the resource type of the delay tag data is identified as "stored energy" and its real-time schedulable capacity (SOC) is greater than 80%, the reference decay factor is increased by 0.2, and if the SOC is less than or equal to 80%, it is decreased by 0.2. The SOC threshold is statistically determined from historical operating data of the energy storage device in the virtual power plant operating log, and analysis of the past 500 scheduling records reveals that when the SOC is greater than 80%, the average adjustment amount of the energy storage device is increased by 20% over the time when the SOC is less than or equal to 80% (e.g., 100kW when the SOC is 85% and 80kW when the SOC is 75%). To reflect the adjustment performance difference, the offset is set to 0.2 (20% improvement in adjustment corresponds to 0.2 improvement in priority). For example, the SOC of a certain energy storage device is 85%, the reference attenuation factor is adjusted from 0.9 to 1.1, and the SOC is adjusted to 0.7 when the SOC is 75%. Non-energy storage type resources (such as photovoltaics and loads) are not subjected to capacity adjustment, and the adjustment capacity of the non-energy storage type resources is not directly related to the capacity.
And taking the ratio of the adjusted reference attenuation factor to the adjustment cost parameter as a dynamic priority compensation coefficient. The adjustment cost parameter is derived from the unit adjustment amount cost (meta/kWh) filled at the time of resource registration, and is written in the "cost_parameter" field of the unified data format. For example, the reference attenuation factor of a certain energy storage device is 1.1, the adjustment cost is 5 yuan/kWh, the compensation coefficient is 1.1/5=0.22, the reference attenuation factor of a certain photovoltaic device is 0.6, the adjustment cost is 3 yuan/kWh, and the compensation coefficient is 0.6/3=0.2. The result of the calculation retains two-bit decimal and writes into the "accounting_coeffective" field for later step invocation.
And carrying out secondary correction on the dynamic priority compensation coefficient according to the power grid node voltage out-of-limit risk level, wherein if the voltage deviation exceeds 5%, for example, the dynamic priority compensation coefficient is judged to be high in risk, the dynamic priority compensation coefficient is multiplied by a suppression factor of 0.5, the deviation is middle risk when the deviation is 3-5%, the deviation is multiplied by 0.8, and the deviation is low risk when the deviation is less than 3%, and the deviation is multiplied by 1.0.
The inhibitor is set by analyzing the contribution degree of the resource regulation to the voltage stabilization in the historical voltage out-of-limit event, for example, in the high risk event that the statistical 50 times of voltage deviation is more than 5%, the contribution degree of the resource regulation to the voltage recovery is reduced by 50% on average, so the inhibitor is set to be 0.5, the contribution degree is reduced by 20% when the deviation is 3% -5%, and the inhibitor is 0.8. For example, the voltage deviation of a certain node is 6%, the compensation coefficient 0.22 is corrected to 0.22×0.5=0.11, and the voltage deviation is corrected to 0.22×0.8=0.176 when the voltage deviation is 4%. The modified coefficients are updated to the "compensation_coeffient" field.
Fusing the initial priority rule and the dynamic priority compensation coefficient to generate a real-time dynamic priority sequence, which comprises the following steps:
Classifying and layering the initial priority score of the non-delay data flow set and the dynamic priority compensation coefficient of the delay data flow set according to the coupling relation between the frequency modulation demand level of the power grid and the resource type identifier;
The layering priority order is dynamically adjusted based on the real-time power grid frequency deviation change direction, wherein energy storage type resources are preferentially called when the frequency is positively deviated, and load type resources are preferentially called when the frequency is negatively deviated;
Carrying out multidimensional cross sequencing on non-delay data and delay data according to the resource type identification and the grid frequency modulation demand level, and generating a priority subsequence of the resource type-frequency modulation level dimension;
The priority subsequence is safely screened according to the power grid node voltage out-of-limit risk level, namely resource items which are likely to exacerbate the voltage problem are removed under the high risk level, and the complete priority subsequence is reserved under the low risk level;
And merging all the priority subsequences after the safety screening, and generating a real-time dynamic priority sequence according to a global priority rule of the grid frequency modulation demand level and the resource type identifier.
And classifying and layering the initial priority score of the non-delayed data flow set and the dynamic priority compensation coefficient of the delayed data flow set according to the coupling relation between the frequency modulation demand level of the power grid and the resource type identifier. The classification and layering rule is that when the frequency modulation demand level of the power grid is urgent, the initial priority grade and the dynamic priority compensation coefficient of the energy storage type resource are classified into a first level, the load type resource is classified into a second level, all resource types under the normal level are classified into the same level according to the initial priority grade and the compensation coefficient numerical value, and the loose level is classified into different levels according to the adjustment cost parameter from low to high. For example, the initial priority score (e.g., 8.0) and the compensation factor (e.g., 0.22) of the energy storage resource at the emergency level are both classified as the first level, and the score (e.g., 5.0) and the factor (e.g., 0.15) of the controllable load are classified as the second level.
The layering priority order is dynamically adjusted based on the real-time power grid frequency deviation change direction, namely, when the frequency deviation is positive (namely, the frequency is higher than the rated value), the energy storage type resources are preferentially called to absorb surplus electric energy, and when the frequency deviation is negative (namely, the frequency is lower than the rated value), the controllable load type resources are preferentially called to cut down the electricity consumption requirement. For example, when the positive deviation of the frequency is detected to be 0.3Hz, the first level subsequence of the energy storage type resource is arranged at the top of the sorting queue, and if the negative deviation of the frequency is detected to be 0.4Hz, the second level subsequence of the load type resource is arranged at the top. The frequency deviation direction is determined by comparing the real-time frequency value with a nominal value (e.g., 50Hz or 60 Hz), and the deviation is calculated by subtracting the nominal value from the current frequency value.
And carrying out multidimensional cross sequencing on the non-delay data and the delay data according to the resource type identification and the grid frequency modulation demand level, and generating a priority subsequence of the resource type-frequency modulation level dimension. The cross ordering rule is that the emergency grade data item is prioritized over the normal grade under the same resource type, the normal grade is prioritized over the loose grade, and the non-delay data is prioritized over the delay data under the same frequency modulation grade. For example, the emergency level non-delay data (initial score 8.0) of the energy storage class resource is ranked before the emergency level delay data (compensation coefficient 0.22), and the normal level non-delay data (score 6.0) is ranked before the normal level delay data (coefficient 0.18). All the subsequences are stored in a JSON array format, and the elements in the array comprise resource identification, priority value and frequency modulation class identification fields.
And carrying out safety screening on the priority subsequence according to the power grid node voltage out-of-limit risk level, wherein if the problem of voltage out-of-limit is possibly aggravated after the scheduling instruction of a certain resource is executed, the resource item is removed from the subsequence under the high risk level. For example, when the voltage deviation of a node is up to 5.2%, a certain energy storage device is judged to be high in risk and rejected if the voltage is further increased to 5.8% by the charge and discharge operation, and if the voltage deviation is 3.5%, the item is reserved. The voltage out-of-limit risk prediction is realized through historical data analysis, the influence trend of the adjustment behavior of similar resources under similar voltage deviation scenes is counted, and if the deviation is enlarged due to more than 80% of cases, the high-risk resources are marked.
And merging all the priority subsequences after the safety screening, and generating a real-time dynamic priority sequence according to a global priority rule of the grid frequency modulation demand level and the resource type identifier. The merging rule is that the whole emergency level sub-sequence is prioritized over the normal level, the normal level is prioritized over the loose level, and the sub-sequences are arranged according to the preset sequence of the resource type identifiers (such as energy storage > photovoltaic > load) under the same frequency modulation level. For example, the emergency level stored energy subsequence is ordered before all normal level subsequences, and the normal level loaded subsequence is ordered after all loose level subsequences. In the finally generated real-time dynamic priority sequence, each piece of data comprises a resource identifier, a priority value, a frequency modulation grade identifier and a security screening state field.
Simulating the influence of different scheduling instruction combinations on the power grid physical safety constraint index based on the real-time dynamic priority sequence, and generating a safety margin assessment result, wherein the method comprises the following steps:
Generating a plurality of candidate scheduling instruction combinations according to the real-time dynamic priority sequence, wherein each candidate scheduling instruction combination comprises scheduling instructions with different resource type identifiers and corresponding adjustment quantity parameters;
Performing power grid physical security constraint simulation on each candidate scheduling instruction combination based on current operation parameters of the power grid, and predicting power grid physical security constraint index change values after the candidate scheduling instruction combination is executed;
comparing the predicted power grid physical security constraint index change value with a preset power grid physical security constraint index security threshold value to generate a security margin evaluation value of each candidate scheduling instruction combination;
and screening candidate scheduling instruction combinations with the safety margin evaluation value larger than or equal to a preset safety margin evaluation threshold value, and generating a safety margin evaluation result set.
The real-time dynamic priority sequence is derived from the priority ordering result generated in the step S4, and candidate scheduling instruction combinations are sequentially selected from high priority to low priority, wherein each combination comprises scheduling instructions with at least two resource type identifiers. For example, the first 5 resource items are selected from the priority sequence, a scheduling instruction combination containing energy storage, photovoltaic and controllable load is generated, the adjustment quantity parameter of the energy storage instruction is a charge-discharge power value (such as 100 kW), the photovoltaic instruction is an output adjustment value (such as 50 kW), and the load instruction is a reduction quantity value (such as 80 kW). The adjustment quantity parameter is set according to the historical adjustment capacity range corresponding to the resource type identifier, the charge and discharge power value of the energy storage device does not exceed 90% of the rated capacity of the energy storage device, and the photovoltaic output adjustment value does not exceed the maximum adjustable quantity under the current illumination condition.
The current operation parameters of the power grid comprise node voltage, line current and power grid frequency real-time values, and the current operation parameters are derived from a real-time data acquisition interface of a power grid monitoring system. The simulation process adopts a tide calculation method based on kirchhoff's law and a power balance equation to predict voltage deviation values of all nodes, load rate changes of lines and power grid frequency deviation change values after a scheduling instruction is executed. For example, after simulation is performed on a certain candidate scheduling command combination, the voltage of the predicted node A rises from 10kV to 10.5kV (deviation 5%), the load rate of the line L1 increases from 80% to 85%, and the power grid frequency decreases from 50.0Hz to 49.8Hz (deviation-0.2 Hz).
And comparing the predicted power grid physical security constraint index change value with a preset power grid physical security constraint index security threshold value to generate a security margin evaluation value of each candidate scheduling instruction combination. The grid physical safety constraint index safety threshold comprises a node voltage deviation safety threshold (such as +/-5%), a line load rate safety threshold (such as 90%), and a grid frequency deviation safety threshold (such as +/-0.5 Hz).
The calculation method of the safety margin evaluation value comprises the steps of taking the percentage of the difference value of the node voltage deviation safety threshold value and the predicted voltage deviation value to the safety threshold value, the percentage of the difference value of the line load rate safety threshold value and the predicted load rate value to the safety threshold value, and the percentage of the difference value of the grid frequency deviation safety threshold value and the predicted frequency deviation value to the safety threshold value, wherein the minimum value of the three values is used as the safety margin evaluation value of the combination. For example, if the predicted voltage deviation of a certain combination is 4% (the difference percentage is (5% -4%)/5% = 20%), the load factor is 88% (the difference percentage is (90% -88%)/90% ≡2.2%), the frequency deviation is-0.3 Hz (the difference percentage is (0.5-0.3)/0.5=40%), the safety margin evaluation value is 2.2%.
And screening candidate scheduling instruction combinations with the safety margin evaluation value larger than or equal to a preset safety margin evaluation threshold value, and generating a safety margin evaluation result set. The preset safety margin evaluation threshold is set according to the lowest margin value of the safety operation in the historical scheduling record, for example, the margin value distribution of the fault-free scheduling instruction in the past year is counted, and 90% of the decimal number (such as 10%) is taken as the threshold. If the safety margin evaluation value of the candidate scheduling instruction combination is smaller than 10%, marking as out-of-limit risk and eliminating, and if the safety margin evaluation value is larger than or equal to 10%, reserving the safety margin evaluation result set. For example, if the safety margin evaluation value of a certain combination is 8%, the combination is rejected, and if the safety margin evaluation value of another combination is 12%, the combination is reserved. Each piece of data in the safety margin evaluation result set comprises a candidate scheduling instruction combination identifier (such as combination number 001), a safety margin evaluation value (such as 12%) and an out-of-limit risk marking field (such as 'safety' or 'out-of-limit') which are used for being called when a final scheduling instruction is generated in a subsequent step.
The physical safety constraint indexes of the power grid comprise node voltage deviation, line load rate and frequency deviation, wherein the node voltage deviation is defined as the difference percentage of the actual voltage of the node and the rated voltage, the line load rate is defined as the ratio percentage of the actual current of the line and the rated current-carrying capacity, and the frequency deviation is defined as the absolute value of the difference value of the actual frequency of the power grid and the rated frequency. The safety threshold is set according to the tolerance limit of the equipment in the historical operation data, for example, the safety threshold of node voltage deviation is rated value + -5% (based on the insulation tolerance capability of a transformer), the threshold of line load rate is 90% (based on the thermal stability limit of the line), and the threshold of frequency deviation is + -0.5 Hz (based on the frequency modulation dead zone of a generator set). The index predicted value is obtained through the power grid state change after the power flow calculation simulation scheduling instruction is executed.
According to the safety margin evaluation result, the real-time dynamic priority sequence is adjusted, a resource scheduling instruction is generated and distributed to corresponding resource execution through a shared communication network, and the method comprises the following steps:
Adjusting the real-time dynamic priority sequence according to the safety margin evaluation value in the safety margin evaluation result set, and if the safety margin evaluation value of the candidate scheduling instruction combination is lower than a preset safety margin evaluation threshold value, eliminating the corresponding resource item from the real-time dynamic priority sequence;
Generating a resource scheduling instruction based on the adjusted real-time dynamic priority sequence, wherein the resource scheduling instruction comprises a resource type identifier, an adjustment quantity parameter and an execution time window;
Distributing the resource scheduling instruction to corresponding distributed energy resource terminal equipment through a shared communication network;
After the distributed energy resource terminal equipment receives the resource scheduling instruction, executing the adjustment quantity parameter according to the control logic corresponding to the resource type identifier, and feeding back an execution result to the virtual power plant aggregation control platform through the shared communication network.
The preset safety margin evaluation threshold is 10%, and the safety margin evaluation threshold is set according to the lowest margin quantile of the safety operation in the history scheduling record, for example, 90% quantile is taken after the margin value distribution of the fault-free scheduling instruction in the past year is counted. For example, if the safety margin evaluation value of a certain candidate scheduling instruction combination is 8%, all resource items contained in the combination are deleted from the real-time dynamic priority sequence, and if the evaluation value is 12%, the resource items are reserved. The real-time dynamic priority sequence adjusting process is carried out in the internal memory of the virtual power plant aggregation control platform, and the adjusted sequence reserves the resource type identification, the adjustment quantity parameter and the priority value field.
The execution time window is set according to the preset data acquisition frequency in step S1, for example, when the data acquisition frequency is once every 5 minutes, the execution time window is set to be a 5-minute interval after the start time point of the next acquisition period. The adjustment quantity parameter is extracted from the priority sequence and converted into a control instruction which can be identified by the equipment, for example, a charge-discharge power instruction of the energy storage equipment is "charge_power=100 kW", and an output power adjustment instruction of the photovoltaic equipment is "active_power=50 kW". The packaging format of the resource scheduling instruction is a JSON character string, and the JSON character string comprises a resource type identifier, a regulating variable parameter, an execution time window and an instruction signature field, wherein the instruction signature is used for verifying the validity of an instruction source.
The distribution protocol of the shared communication network matches a preset communication interface and a data encapsulation format according to the resource type identification. For example, when the resource type is identified as 'energy storage', modbus TCP protocol is adopted and distributed through a specific IP port, the data encapsulation format is hexadecimal code stream, and when the resource type is identified as 'photovoltaic', MQTT protocol is adopted and distributed through a topic subscription mechanism, and the data encapsulation format is JSON coded by UTF-8. The communication interface and the data format are predefined in the resource registration stage and stored in a communication configuration database of the virtual power plant aggregation control platform. In the distribution process, the aggregation control platform queries a database according to the resource type identification and sends an instruction.
The control logic comprises charge and discharge power control of the energy storage equipment, inverter output regulation of the photovoltaic equipment and switching value control of the load equipment. For example, after the energy storage device analyzes "charge_power=100 kW", the output power of the converter is adjusted to the target value, and after the photovoltaic device analyzes "active_power=50 kW", the maximum output power of the inverter is limited to 50kW. The execution result comprises a resource identifier, an actual adjustment value, an execution state (success/failure) and a timestamp, is packaged into a JSON format and is returned through the original communication protocol. For example, the energy storage device returns "{'resource_id': 'ESS001', 'actual_power': 98kW, 'status': 'success', 'timestamp': '2023-10-05T08:00:00Z'}". feedback data to be stored in a historical execution log database of the virtual power plant aggregation control platform for use in subsequent scheduling optimization and anomaly analysis.
Embodiment 2 fig. 2 shows a schematic structural diagram of a data sharing system based on a virtual power plant, and the data sharing system based on the virtual power plant includes:
The real-time acquisition module is used for acquiring real-time operation data of the distributed energy resources of the virtual power plant through a shared communication network, wherein the real-time operation data comprise resource type identification and adjustment cost parameters;
The delay marking module is used for judging the transmission delay state of the real-time operation data in the shared communication network and generating a data stream set of a delay mark and a non-delay mark;
The dynamic compensation module is used for matching a preset initial priority rule based on the resource type identification aiming at the non-delay marked data, determining an attenuation factor according to the resource type identification and the grid frequency modulation demand level in a cooperative manner, and generating a dynamic priority compensation coefficient based on the adjustment cost parameter and the attenuation factor aiming at the delay marked data;
the sequence generation module is used for generating a real-time dynamic priority sequence by fusing the initial priority rule and the dynamic priority compensation coefficient;
The safety evaluation module simulates the influence of different dispatching instruction combinations on the physical safety constraint index of the power grid based on the real-time dynamic priority sequence, and generates a safety margin evaluation result;
And the instruction distribution module is used for adjusting the real-time dynamic priority sequence according to the safety margin evaluation result, generating a resource scheduling instruction and distributing the resource scheduling instruction to corresponding resource execution through the shared communication network.
Embodiment 3 a data sharing device based on a virtual power plant includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor, wherein the program or instruction when executed by the processor implements a data sharing method based on the virtual power plant.
Embodiment 4a data sharing medium based on a virtual power plant, on which a program or an instruction is stored, which when executed by a processor implements a data sharing method based on a virtual power plant.
The above formulas are all the dimensionality removal and numerical calculation, the formulas are one formulas which are obtained by collecting a large amount of data and performing software simulation to obtain the closest actual situation, and preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.
It should be noted that the present invention can be deployed on the device itself to implement embedded applications, and also can run on a PC end or other terminals with user interfaces, so as to satisfy various hardware environments and use requirements.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described system, apparatus and module may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or modules, which may be in electrical, mechanical, or other forms.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, may be located in one place, or may be distributed over multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Finally, the foregoing description of the preferred embodiment of the invention is provided for the purpose of illustration only, and is not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.