技术领域Technical Field
本发明属于排污企业生产状态监控技术领域，涉及基于用电数据的管控状态下排污企业停或限产监测方法。The present invention belongs to the technical field of production status monitoring of pollutant-discharging enterprises, and relates to a method for monitoring the suspension or limitation of production of pollutant-discharging enterprises under a control state based on electricity consumption data.
背景技术Background technique
在实际生产中，由于环境控制、电力供应、产业管理等限制因素，相关部门会定向的对企业在规定期限内提出停产、限产的要求，并对企业的停/限产执行情况进行监管。由于企业数量多，行业类别多，工艺错综复杂，判断污染源耗费精力；而监察人员有限，监察力度要求又非常大，人工排查耗时耗力，难免有疏漏，精确度不够，使得管理制度效能难以充分发挥。In actual production, due to restrictive factors such as environmental control, power supply, and industrial management, relevant departments will specifically require enterprises to stop or limit production within a specified period, and supervise the implementation of the suspension/limitation of production. Due to the large number of enterprises, many industry categories, and complex processes, it takes a lot of effort to determine the source of pollution; while the number of supervisors is limited and the supervision intensity requirements are very high, manual investigation is time-consuming and labor-intensive, and it is inevitable that there will be omissions and insufficient accuracy, making it difficult to give full play to the effectiveness of the management system.
针对管控背景下，“是否可以利用企业用电数据，对企业的停/限产行为进行精准研判”，这是本发明着手解决的问题。In the context of control, "whether it is possible to use the electricity consumption data of enterprises to accurately judge the suspension/production restriction behavior of enterprises" is the problem that the present invention aims to solve.
用电数据客具有观准确、不可篡改的特性，并且其能够与企业运行状况紧密相连，可用于企业的停/限产行为的分析研判。Electricity consumption data is objective, accurate, and cannot be tampered with. It is closely linked to the operating conditions of the enterprise and can be used to analyze and judge the enterprise's suspension/production restriction behavior.
一般来说，管控状态下，排污企业由于处于停产和限产状态，与常态情况下相比的，企业的用电量应该是减少的。但是由于企业每日用电量存在一定随机波动性，管控状态下减少的电量是由于随机波动所产生的，还是由于采取了停产或限产措施导致的，则需要进行进一步的量化分析。Generally speaking, under the control state, the electricity consumption of polluting enterprises should be reduced compared with normal conditions due to the suspension and production restriction of polluting enterprises. However, due to the random fluctuation of daily electricity consumption of enterprises, whether the reduction of electricity under the control state is caused by random fluctuation or due to the suspension or production restriction measures, further quantitative analysis is needed.
发明内容Summary of the invention
为解决现有技术中的不足，本申请提供基于用电数据的管控状态下排污企业停或限产监测方法。In order to address the deficiencies in the prior art, the present application provides a method for monitoring the shutdown or production restriction of pollutant-discharging enterprises under a control state based on electricity consumption data.
为了实现上述目标，本发明采用如下技术方案：In order to achieve the above objectives, the present invention adopts the following technical solutions:
基于用电数据的管控状态下排污企业停或限产监测方法，其特征在于：The method for monitoring the shutdown or production restriction of pollutant-discharging enterprises under the control state based on electricity consumption data is characterized by:
所述方法包括以下步骤：The method comprises the following steps:
步骤1：获取待监测污水企业的历史用电数据，及相应的环境部门对管控状态下停限产违规企业的处罚数据；Step 1: Obtain the historical electricity consumption data of the sewage enterprises to be monitored, and the penalty data of the corresponding environmental departments on the enterprises that violated the regulations and were suspended or restricted from production under the control status;
步骤2：对待监测污水企业的历史用电数据进行用电指标计算，得到用电指标样本数据；Step 2: Calculate the electricity consumption index based on the historical electricity consumption data of the sewage enterprises to be monitored, and obtain the electricity consumption index sample data;
步骤3：结合环境部门对管控状态下停限产违规企业的处罚数据，从用电指标样本数据中筛选状态判别变量样本数据；Step 3: Combined with the penalty data of environmental departments on enterprises that violated the production control regulations and suspended or restricted production, the state discriminant variable sample data is selected from the electricity consumption index sample data;
步骤4：基于状态判别变量样本数据，采用逻辑回归构建管控状态下生产状态监测模型；Step 4: Based on the sample data of the state discriminant variable, a production state monitoring model under the control state is constructed using logistic regression;
步骤5：管控状态下，实时获取待监测污水企业的用电数据，进行用电指标计算后输入管控状态下生产状态监测模型，得到企业是否执行管控状态下的停限产监测结果。Step 5: Under the control state, obtain the electricity consumption data of the sewage enterprise to be monitored in real time, calculate the electricity consumption index and input it into the production status monitoring model under the control state to obtain the monitoring results of whether the enterprise implements the production suspension and restriction under the control state.
本发明进一步包括以下优选方案：The present invention further includes the following preferred embodiments:
优选地，步骤1中，通过环境部门提供的排污企业监测清单，筛选获取各个行业中企业生产与电力消耗关系密切的排污企业作为可监测排污企业；Preferably, in step 1, the monitoring list of pollutant-discharging enterprises provided by the environmental department is used to screen and obtain pollutant-discharging enterprises in various industries whose production and electricity consumption are closely related as monitorable pollutant-discharging enterprises;
对待监测排污企业进行监测前，预先判断其是否为可监测排污企业，若是，则可进行监测，若不是，则退出并发出不可监测提示。Before monitoring a pollutant-discharging enterprise, determine in advance whether it is a monitorable pollutant-discharging enterprise. If so, monitoring can be carried out. If not, exit and issue a non-monitorable prompt.
优选地，通过能源结构中电力占比度量企业生产与电力消耗关系密切度，筛选能源结构中电力占比超过80％的企业作为可监测排污企业。Preferably, the closeness of the relationship between enterprise production and electricity consumption is measured by the proportion of electricity in the energy structure, and enterprises with a proportion of electricity exceeding 80% in the energy structure are screened as monitorable pollution-discharging enterprises.
优选地，步骤1中，确定待监测污水企业及其管控时间，获取管控前一周待监测污水企业的用电数据，及相应的环境部门对管控状态下停限产违规企业的处罚数据；Preferably, in step 1, the sewage enterprises to be monitored and their control time are determined, and the electricity consumption data of the sewage enterprises to be monitored in the week before the control is obtained, as well as the penalty data of the corresponding environmental departments on the enterprises that violated the regulations and suspended or restricted production under the control state;
所述用电数据具体包括：日期、排污企业当日总用电量、24小时各时段用电量；The electricity consumption data specifically include: date, total electricity consumption of the pollutant-discharging enterprise on that day, and electricity consumption in each period of 24 hours;
所述处罚数据具体包括：处罚日期、处罚原因、违法日期、处罚前后用电数据。The penalty data specifically includes: penalty date, penalty reason, violation date, and electricity usage data before and after the penalty.
优选地，所述用电指标包括企业违规管控前7天日平均用电量x1、企业违规管控前7天日用电量样本标准差x2、管控当天企业用电量与管控前一周平均用电量的比值x3、企业管控当天节假日变量x4，具体的计算公式为：Preferably, the electricity consumption index includes the daily average electricity consumption x1 of the enterprise 7 days before the illegal control, the daily electricity consumption sample standard deviation x2 of the enterprise 7 days before the illegal control, the ratio of the enterprise electricity consumption on the day of control to the average electricity consumption in the week before the control x3 , and the holiday variable x4 on the day of enterprise control. The specific calculation formula is:
企业违规管控前7天日平均用电量x1：Average daily electricity consumption of the enterprise in the 7 days before the violation control x1 :
zt为管控前一天日用电量；zt is the daily electricity consumption on the previous day of control;
企业违规管控前7天日用电量样本标准差x2：Standard deviation of daily electricity consumption samples of enterprises in the 7 days before the violation control x2 :
管控当天企业用电量与管控前一周平均用电量的比值x3：The ratio of the enterprise's electricity consumption on the day of control to the average electricity consumption in the week before control x3 :
zt+1表示管控当天用电量。zt+1 means controlling the electricity consumption on that day.
企业管控当天节假日变量x4：节假日取1，工作日取0。The variable x4 for the holiday on the day of enterprise control is 1 for holidays and 0 for weekdays.
优选地，步骤3中，引入状态判别变量y表示企业是否执行管控状态下的停限产；Preferably, in step 3, a state discrimination variable y is introduced to indicate whether the enterprise implements the production suspension or restriction under the control state;
所述状态判别变量y为两值型变量，y＝0表示管控状态下按照要求进行停限产，y＝1表示未要按照要求停限产。The state discrimination variable y is a two-value variable, y=0 indicates that production is suspended or restricted as required under the control state, and y=1 indicates that production is not suspended or restricted as required.
优选地，步骤3具体为：Preferably, step 3 is specifically:
以环境部门对管控状态下停限产违规企业的处罚数据中的企业违法日期为基期，从步骤2的用电指标样本数据中获取企业违法日期当天及前7天的用电数据计算所得的用电指标，作为状态判别变量y＝1的样本数据；Taking the enterprise violation date in the penalty data of the environmental department on the enterprises that violated the regulations and were suspended or restricted from production under control as the base period, the electricity consumption index calculated from the electricity consumption data on the enterprise violation date and the previous 7 days is obtained from the electricity consumption index sample data in step 2 as the sample data of the state discriminant variable y=1;
以重要节假日为管控期，从步骤2的用电指标样本数据中获取重要节假日前7天的用电数据计算得到的用电指标，作为状态判别变量y＝0的样本数据；Taking important holidays as the control period, the electricity consumption index calculated by obtaining the electricity consumption data 7 days before important holidays from the electricity consumption index sample data in step 2 is used as the sample data of the state discriminant variable y=0;
所述重要节假日为假期超过2日的节假日。The important holidays are holidays with a duration of more than 2 days.
优选地，步骤4中，以构建状态判别变量y＝1为因变量，各用电指标为自变量建立关于y＝1概率p和自变量的logistic回归模型，采用状态判别变量样本数据，确定模型的回归系数，得到管控状态下生产状态监测模型。Preferably, in step 4, a logistic regression model about the probability p of y=1 and the independent variable is established by constructing the state discriminant variable y=1 as the dependent variable and each electricity consumption index as the independent variable, and the regression coefficient of the model is determined by using the state discriminant variable sample data to obtain the production status monitoring model under the control state.
优选地，所述logistic回归模型为：Preferably, the logistic regression model is:
Logit(p)＝α0+α1x1+α2x2+α3x3+α4x4Logit(p )＝α0 +α1x1+α2x2+α3x3+α4x4
其中，x1为企业违规管控前7天日平均用电量、x2为企业违规管控前7天日用电量样本标准差、x3为管控当天企业用电量与管控前一周平均用电量的比值x3、x4为企业管控当天节假日变量；α0，α1，α2，α3和α4表示回归系数。Among them,x1 is the average daily electricity consumption of the enterprise 7 days before the illegal control,x2 is the sample standard deviation of daily electricity consumption 7 days before the illegal control,x3 is the ratio of the enterprise's electricity consumption on the day of control to the average electricity consumption in the week before the control,andx4 is the holiday variable on the day of enterprise control;α0 ,α1 ,α2 ,α3 andα4 represent regression coefficients.
优选地，步骤5中，管控状态下，实时获取待监测污水企业的用电数据并计算用电指标，将其带入监测模型，得到管控状态下排污企业异常生产的概率预测值；Preferably, in step 5, under the control state, the electricity consumption data of the sewage enterprise to be monitored is obtained in real time and the electricity consumption index is calculated, which is brought into the monitoring model to obtain the probability prediction value of abnormal production of the sewage discharge enterprise under the control state;
同时设置概率阈值，当预测概率值大于概率阈值，将状态判别变量y预测值取为1，否则取为0；At the same time, the probability threshold is set. When the predicted probability value is greater than the probability threshold, the predicted value of the state discriminant variable y is taken as 1, otherwise it is taken as 0;
y＝0表示管控状态下按照要求进行停限产，y＝1表示未要按照要求停限产。y=0 means that production is suspended or restricted as required under the control status, and y=1 means that production is not suspended or restricted as required.
本申请所达到的有益效果：Beneficial effects achieved by this application:
本发明利用用电数据的不可篡改特性，分析排污企业中个体与总体、异常与常态等模式下的用电特征，确定用电指标，结合历史用电数据，及相应的环境部门对管控状态下停限产违规企业的处罚数据，构建管控状态下生产状态监测模型，实现排污企业异常行为的侦测和预警，能够更加客观地反映出排污的企业的生产状况，实现对企业是否执行管控状态下的停限产情况的自动监控，节省大量的人力物力。The present invention utilizes the non-tamperable property of electricity consumption data to analyze the electricity consumption characteristics of individual and overall, abnormal and normal modes in polluting enterprises, determine electricity consumption indicators, and combines historical electricity consumption data with the penalty data of corresponding environmental departments on enterprises that violate the regulations and have suspended or restricted production under the control status to construct a production status monitoring model under the control status, thereby realizing the detection and early warning of abnormal behavior of polluting enterprises, and being able to more objectively reflect the production status of polluting enterprises, realize automatic monitoring of whether enterprises implement the suspension or restriction of production under the control status, thus saving a lot of manpower and material resources.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1是本发明方法流程图；Fig. 1 is a flow chart of the method of the present invention;
图2是本发明实施例中的样本数据；FIG2 is sample data in an embodiment of the present invention;
图3是本发明实施例中的概率预测结果。FIG. 3 is a probability prediction result in an embodiment of the present invention.
具体实施方式Detailed ways
下面结合附图对本申请作进一步描述。以下实施例仅用于更加清楚地说明本发明的技术方案，而不能以此来限制本申请的保护范围。The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present application.
如图1所示，本发明的基于用电数据的管控状态下排污企业停或限产监测方法，包括以下步骤：As shown in FIG1 , the method for monitoring the shutdown or production restriction of pollutant-discharging enterprises under the control state based on electricity consumption data of the present invention comprises the following steps:
步骤1：获取待监测污水企业的历史用电数据，及相应的环境部门对管控状态下停限产违规企业的处罚数据；Step 1: Obtain the historical electricity consumption data of the sewage enterprises to be monitored, and the penalty data of the corresponding environmental departments on the enterprises that violated the regulations and were suspended or restricted from production under the control status;
进一步的，具体实施时，通过环境部门提供的排污企业监测清单，采用分层抽样方法筛选获取各个行业中企业生产与电力消耗关系密切的排污企业作为可监测排污企业；Furthermore, during the specific implementation, through the monitoring list of pollutant-discharging enterprises provided by the environmental department, a stratified sampling method is used to select pollutant-discharging enterprises in various industries that have a close relationship between production and electricity consumption as monitorable pollutant-discharging enterprises;
对待监测排污企业进行监测前，预先判断其是否为可监测排污企业，若是，则可进行监测，若不是，则退出并发出不可监测提示。Before monitoring a pollutant-discharging enterprise, determine in advance whether it is a monitorable pollutant-discharging enterprise. If so, monitoring can be carried out. If not, exit and issue a non-monitorable prompt.
通过能源结构中电力占比度量企业生产与电力消耗关系密切度，筛选能源结构中电力占比超过80％的企业作为可监测排污企业。The closeness of the relationship between enterprise production and electricity consumption is measured by the proportion of electricity in the energy structure, and enterprises whose electricity proportion in the energy structure exceeds 80% are selected as monitorable pollution-discharging enterprises.
步骤1中，确定待监测污水企业及其管控时间，获取管控前一周待监测污水企业的用电数据，及相应的环境部门对管控状态下停限产违规企业的处罚数据；In step 1, determine the sewage enterprises to be monitored and their control time, obtain the electricity consumption data of the sewage enterprises to be monitored in the week before the control, and the penalty data of the corresponding environmental departments on the enterprises that violated the control and suspension of production;
所述用电数据具体包括：日期、排污企业当日总用电量、24小时各时段用电量；The electricity consumption data specifically include: date, total electricity consumption of the pollutant-discharging enterprise on that day, and electricity consumption in each period of 24 hours;
所述处罚数据具体包括：处罚日期、处罚原因、违法日期、处罚前后用电数据。The penalty data specifically includes: penalty date, penalty reason, violation date, and electricity usage data before and after the penalty.
步骤2：对待监测污水企业的历史用电数据进行用电指标计算，得到用电指标样本数据；Step 2: Calculate the electricity consumption index based on the historical electricity consumption data of the sewage enterprises to be monitored, and obtain the electricity consumption index sample data;
由于企业每天用电量数值呈现出一定随机波动，管控前后的由于排污企业被停限产，用电数量上面也会系统性波动，两种波动的叠加，对于排污企业是否按照管控要求进行停限产的研判带来了困难。为了能够区分出随机波动和系统性波动，本发明具体实施时选取的用电指标包括企业违规管控前7天日平均用电量x1、企业违规管控前7天日用电量样本标准差x2、管控当天企业用电量与管控前一周平均用电量的比值x3、企业管控当天节假日变量x4；Since the daily electricity consumption of enterprises shows certain random fluctuations, the electricity consumption will also fluctuate systematically due to the suspension and restriction of production of polluting enterprises before and after the control. The superposition of the two fluctuations makes it difficult to judge whether the polluting enterprises have suspended and restricted production in accordance with the control requirements. In order to distinguish between random fluctuations and systematic fluctuations, the electricity consumption indicators selected in the specific implementation of the present invention include the daily average electricity consumption of the enterprise 7 days before the illegal control x1 , the daily electricity consumption sample standard deviation of the enterprise 7 days before the illegal control x2 , the ratio of the enterprise's electricity consumption on the day of control to the average electricity consumption in the week before the control x3 , and the holiday variable x4 on the day of the enterprise's control;
具体的计算公式为：The specific calculation formula is:
企业违规管控前7天日平均用电量x1：Average daily electricity consumption of the enterprise in the 7 days before the violation control x1 :
zt为管控前一天日用电量；zt is the daily electricity consumption on the previous day of control;
企业违规管控前7天日用电量样本标准差x2：Standard deviation of daily electricity consumption samples of enterprises in the 7 days before the violation control x2 :
管控当天企业用电量与管控前一周平均用电量的比值x3：The ratio of the enterprise's electricity consumption on the day of control to the average electricity consumption in the week before control x3 :
zt+1表示管控当天用电量。zt+1 means controlling the electricity consumption on that day.
企业管控当天节假日变量x4：节假日取1，工作日取0。The variable x4 for the holiday on the day of enterprise control is 1 for holidays and 0 for weekdays.
进一步的，所述节假日包括周六、周日或国家法定假期如国庆节、中秋节等。Furthermore, the holidays include Saturdays, Sundays or national statutory holidays such as National Day, Mid-Autumn Festival, etc.
步骤3：结合环境部门对管控状态下停限产违规企业的处罚数据，从用电指标样本数据中筛选状态判别变量样本数据；Step 3: Combined with the penalty data of environmental departments on enterprises that violated the production control regulations and suspended or restricted production, the state discriminant variable sample data is selected from the electricity consumption index sample data;
具体实施时，引入状态判别变量y表示企业是否执行管控状态下的停限产；In specific implementation, the state discrimination variable y is introduced to indicate whether the enterprise implements the suspension and restriction of production under the control state;
所述状态判别变量y为两值型变量，y＝0表示管控状态下按照要求进行停限产，y＝1表示未要按照要求停限产。The state discrimination variable y is a two-value variable, y=0 indicates that production is suspended or restricted as required under the control state, and y=1 indicates that production is not suspended or restricted as required.
为了获取到能够训练判别模型的真实样本数据。In order to obtain real sample data that can train the discriminant model.
对于y＝1的确定，根据环保部门对于排污企业的处罚数据进行处理而得。例如：根据环境保护局提供2020年排污企业管控状态下违法处罚清单99条，以企业违法日期为基期，获取当天及前7天的用电数据，并计算x1-x4,相应的状态判别变量y＝1。The determination of y=1 is based on the penalty data of the environmental protection department on polluting enterprises. For example, according to the 99 illegal penalty lists of polluting enterprises under the control status in 2020 provided by the Environmental Protection Bureau, the date of the enterprise's violation is taken as the base period, the electricity consumption data of the day and the previous 7 days are obtained, andx1-x4 is calculated, and the corresponding state discrimination variable y=1.
对于y＝0的确定，通过数据的探索性分析发现，在重要节假日(如：春节、国庆、五一、中秋等)期间，由于放假导致企业用电呈现出一个明显的波动，而这样的波动也是符合管控状态下企业停限产的特征。因此，选取重大节假日前后排污企业的用电数据，以节假日视为管控期，将其前7天的数据计算x1-x4,将这些样本数据的状态判别变量取值为y＝0。As for the determination of y=0, through exploratory analysis of data, it is found that during important holidays (such as Spring Festival, National Day, May Day, Mid-Autumn Festival, etc.), the electricity consumption of enterprises shows an obvious fluctuation due to holidays, and such fluctuations are also in line with the characteristics of suspension and restriction of production of enterprises under control. Therefore, the electricity consumption data of polluting enterprises before and after major holidays are selected, and the holidays are regarded as the control period. The data of the first 7 days are calculated asx1 -x4 , and the state discriminant variable of these sample data is set to y=0.
具体实施时，步骤3具体为：In specific implementation, step 3 is as follows:
对照环保部门提供的排污企业的处罚清单，以环境部门对管控状态下停限产违规企业的处罚数据中的企业违法日期为基期，从步骤2的用电指标样本数据中获取企业违法日期当天及前7天的用电数据计算所得的用电指标,作为状态判别变量y＝1的样本数据；According to the penalty list of pollutant-discharging enterprises provided by the environmental protection department, the penalty data of the environmental department on the enterprises that violated the regulations and suspended production under the control status are used as the base period. The electricity consumption index calculated from the electricity consumption data on the day of the enterprise's violation date and the previous 7 days is obtained from the electricity consumption index sample data in step 2 as the sample data of the state discriminant variable y = 1;
以重要节假日为管控期，从步骤2的用电指标样本数据中获取重要节假日前7天的用电数据计算得到的用电指标,作为状态判别变量y＝0的样本数据；Taking important holidays as the control period, the electricity consumption index calculated from the electricity consumption data of the 7 days before the important holidays is obtained from the electricity consumption index sample data in step 2, and used as the sample data of the state discrimination variable y=0;
所述重要节假日为假期超过2日的节假日。The important holidays are holidays with a duration of more than 2 days.
步骤4：基于状态判别变量样本数据，采用逻辑回归构建管控状态下生产状态监测模型，具体的：Step 4: Based on the sample data of the state discriminant variable, a logistic regression is used to construct a production state monitoring model under the control state. Specifically:
以构建状态判别变量y＝1为因变量，各用电指标为自变量建立关于y＝1概率p和自变量的logistic回归模型；A logistic regression model about the probability p of y=1 and independent variables is established with the construction state discriminant variable y=1 as the dependent variable and each electricity consumption index as the independent variable;
进一步的，采用状态判别变量样本数据，运用极大似然估计方法进行运算确定模型的回归系数，得到管控状态下生产状态监测模型。Furthermore, the state discriminant variable sample data is used, and the maximum likelihood estimation method is used to calculate and determine the regression coefficient of the model, and the production state monitoring model under the control state is obtained.
所述logistic回归模型为：The logistic regression model is:
Logit(p)＝α0+α1x1+α2x2+α3x3+α4x4Logit(p )＝α0 +α1x1+α2x2+α3x3+α4x4
其中，x1为企业违规管控前7天日平均用电量、x2为企业违规管控前7天日用电量样本标准差、x3为管控当天企业用电量与管控前一周平均用电量的比值x3、x4为企业管控当天节假日变量；α0，α1，α2，α3和α4表示回归系数。Among them,x1 is the average daily electricity consumption of the enterprise 7 days before the illegal control,x2 is the sample standard deviation of daily electricity consumption 7 days before the illegal control,x3 is the ratio of the enterprise's electricity consumption on the day of control to the average electricity consumption in the week before the control,andx4 is the holiday variable on the day of enterprise control;α0 ,α1 ,α2 ,α3 andα4 represent regression coefficients.
具体实施时，按照上述方法得到“y＝0”和“y＝1”的两组状态判别变量样本数据，作为样本数据。In specific implementation, two groups of state discrimination variable sample data of "y=0" and "y=1" are obtained according to the above method as sample data.
将y作为目标变量，则其取值为0和1的两分类变量，而自变量即有连续变量，满足Logistic回归分析要求。If y is used as the target variable, it is a categorical variable with values of 0 and 1, while the independent variable is a continuous variable, which meets the requirements of Logistic regression analysis.
因变量y与自变量x1，x2，x3，x4之间的函数关系可以表示成以下的形式：The functional relationship between the dependent variable y and the independent variables x1 , x2 , x3 , x4 can be expressed in the following form:
其中，α0，α1，α2，α3和α4表示回归系数。通过适当的数学变化，可以写成关于异常(y＝1)概率p和自变量如下的logistic回归模型。Wherein, α0 , α1 , α2 , α3 and α4 represent regression coefficients. Through appropriate mathematical changes, the following logistic regression model can be written with respect to the abnormal (y=1) probability p and the independent variable.
Logit(p)＝α0+α1x1+α2x2+α3x3+α4x4Logit(p )＝α0 +α1x1+α2x2+α3x3+α4x4
其中in
对于模型中回归系数，可以利用前面得到的样本数据，运用极大似然估计方法进行运算而得。The regression coefficients in the model can be calculated using the sample data obtained previously and the maximum likelihood estimation method.
例如：针对排污企业数据，得到部分样本数据如图2所示。For example, for the pollutant-discharging enterprise data, some sample data are obtained as shown in Figure 2.
计算所得监测模型为：The calculated monitoring model is:
Logit(p)＝-2.58-3.06×10-7×x1+1.47×10-7x2+3.16×x3-5.69×10-9×x4Logit(p)＝-2.58-3.06×10-7 ×x1 +1.47×10-7x2 +3.16×x3-5.69 ×10-9 ×x4
步骤5：管控状态下，实时获取待监测污水企业的用电数据，进行用电指标计算后输入管控状态下生产状态监测模型，得到企业是否执行管控状态下的停限产监测结果，具体的：Step 5: Under the control state, obtain the electricity consumption data of the sewage enterprise to be monitored in real time, calculate the electricity consumption index and input it into the production status monitoring model under the control state to obtain the monitoring results of whether the enterprise implements the suspension and restriction of production under the control state. Specifically:
管控状态下，实时获取待监测污水企业的用电数据并计算用电指标，将其带入监测模型，得到管控状态下排污企业异常生产的概率预测值；Under the control state, the electricity consumption data of the sewage enterprises to be monitored are obtained in real time and the electricity consumption index is calculated, which is then brought into the monitoring model to obtain the probability prediction value of abnormal production of the sewage discharge enterprises under the control state;
同时设置概率阈值，当预测概率值大于概率阈值，将状态判别变量y预测值取为1，否则取为0；At the same time, the probability threshold is set. When the predicted probability value is greater than the probability threshold, the predicted value of the state discriminant variable y is taken as 1, otherwise it is taken as 0;
y＝0表示管控状态下按照要求进行停限产，y＝1表示未要按照要求停限产。y=0 means that production is suspended or restricted as required under the control status, and y=1 means that production is not suspended or restricted as required.
对于一个企业，要判断在管控期内是否异常，可以通过收集管控当日及前7天的用电数据，然后运用上述方法计算x1-x4，将其带入监测模型，则得到预测值p，也就是管控状态下企业异常(y＝1)的概率值，如图3所示。For an enterprise, to determine whether it is abnormal during the control period, we can collect the electricity consumption data on the control day and the previous 7 days, and then use the above method to calculate x1 -x4 , and bring it into the monitoring model to obtain the predicted value p, that is, the probability value of the enterprise being abnormal (y=1) under the control state, as shown in Figure 3.
由于此时得到的是y＝1的预测值，而不是最终决策值。进一步通过对概率设定阈值，当预测概率值大于阈值，将y预测值取为1；当预测值小于等于阈值，将y的预测值取为0。Since the predicted value obtained at this time is y=1, rather than the final decision value, we can further set a threshold for the probability. When the predicted probability value is greater than the threshold, the predicted value of y is taken as 1; when the predicted value is less than or equal to the threshold, the predicted value of y is taken as 0.
例如：当设定阈值为0.5，则当预测概率高于0.5的时候，可以认为管控状态下排污企业存在异常情况。For example: when the threshold is set to 0.5, when the predicted probability is higher than 0.5, it can be considered that there are abnormal conditions in the pollutant-discharging enterprises under the control state.
本发明现在已经国网系统中架构号分析平台，针对环境部门给出管控企业清单，针对给定的管控时间，运用所得模型进行预测，给出了疑似异常清单。通过实际验证，可有效提高预测准确率。The present invention has now built an analysis platform for the national grid system, which provides a list of controlled enterprises for the environmental department, and uses the obtained model to make predictions for a given control time, and provides a list of suspected anomalies. Through actual verification, the prediction accuracy can be effectively improved.
本发明申请人结合说明书附图对本发明的实施示例做了详细的说明与描述，但是本领域技术人员应该理解，以上实施示例仅为本发明的优选实施方案，详尽的说明只是为了帮助读者更好地理解本发明精神，而并非对本发明保护范围的限制，相反，任何基于本发明的发明精神所作的任何改进或修饰都应当落在本发明的保护范围之内。The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, and it is not a limitation on the protection scope of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111388973.3ACN113988711B (en) | 2021-11-22 | 2021-11-22 | Pollution discharge enterprise stop or limit production monitoring method based on electricity consumption data in control state |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111388973.3ACN113988711B (en) | 2021-11-22 | 2021-11-22 | Pollution discharge enterprise stop or limit production monitoring method based on electricity consumption data in control state |
| Publication Number | Publication Date |
|---|---|
| CN113988711A CN113988711A (en) | 2022-01-28 |
| CN113988711Btrue CN113988711B (en) | 2024-07-02 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111388973.3AActiveCN113988711B (en) | 2021-11-22 | 2021-11-22 | Pollution discharge enterprise stop or limit production monitoring method based on electricity consumption data in control state |
| Country | Link |
|---|---|
| CN (1) | CN113988711B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114943409A (en)* | 2022-04-11 | 2022-08-26 | 国网山东省电力公司滨州供电公司 | Environment monitoring and analyzing method and system based on electric power data |
| CN114693166A (en)* | 2022-04-22 | 2022-07-01 | 北京中科三清环境技术有限公司 | Method and device for monitoring execution condition of emergency emission reduction measure on line and electronic equipment |
| CN115879726B (en)* | 2022-12-23 | 2023-10-31 | 中国环境科学研究院 | A method for monitoring enterprise production status and emissions |
| CN115965284A (en)* | 2022-12-27 | 2023-04-14 | 国网四川省电力公司电力科学研究院 | Enterprise environment-friendly and environment-friendly measure monitoring method |
| CN117609926B (en)* | 2024-01-23 | 2024-04-16 | 中科三清科技有限公司 | Pollution discharge mechanism production state determining method and device based on power data |
| CN119622526B (en)* | 2025-02-11 | 2025-05-13 | 生态环境部环境工程评估中心 | Stop and limit production identification method, device, equipment and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113095953A (en)* | 2021-03-29 | 2021-07-09 | 福建亿力优能电力科技有限公司 | Enterprise environment-friendly monitoring method based on electricity consumption data |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764584B (en)* | 2018-06-05 | 2021-04-09 | 国网浙江省电力有限公司 | Enterprise electric energy substitution potential evaluation method |
| CN110889088B (en)* | 2019-11-04 | 2023-10-20 | 国网浙江省电力有限公司信息通信分公司 | Enterprise pollution discharge supervision method assisted by electric power model |
| CN111523794B (en)* | 2020-04-21 | 2020-11-24 | 国网四川省电力公司电力科学研究院 | Research and judgment method of environmental protection management and control measures based on electricity consumption characteristics of polluting enterprises |
| CN111522864B (en)* | 2020-04-21 | 2020-11-10 | 国网四川省电力公司电力科学研究院 | Enterprise production mode recognition and transfer production early warning method based on electricity consumption data |
| CN111210170B (en)* | 2020-04-21 | 2020-07-31 | 国网四川省电力公司电力科学研究院 | Environment-friendly management and control monitoring and evaluation method based on 90% electricity distribution characteristic index |
| CN111524032B (en)* | 2020-04-21 | 2020-11-10 | 国网四川省电力公司电力科学研究院 | Environment-friendly response quantification method and device based on enterprise electricity consumption data |
| CN112307435B (en)* | 2020-10-30 | 2024-05-31 | 三峡大学 | Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend |
| CN112949700A (en)* | 2021-02-19 | 2021-06-11 | 国网北京市电力公司 | Method and device for identifying execution strength of enterprise yield limit policy |
| CN113298422B (en)* | 2021-06-16 | 2023-01-31 | 福建亿力优能电力科技有限公司 | Pollution source enterprise illegal production monitoring method based on electricity consumption data |
| CN113495912A (en)* | 2021-07-05 | 2021-10-12 | 晟至技术有限公司 | Scattered pollution enterprise accurate prevention and control monitoring method and system based on electric power data |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113095953A (en)* | 2021-03-29 | 2021-07-09 | 福建亿力优能电力科技有限公司 | Enterprise environment-friendly monitoring method based on electricity consumption data |
| Publication number | Publication date |
|---|---|
| CN113988711A (en) | 2022-01-28 |
| Publication | Publication Date | Title |
|---|---|---|
| CN113988711B (en) | Pollution discharge enterprise stop or limit production monitoring method based on electricity consumption data in control state | |
| CN111310803B (en) | Environment data processing method and device | |
| US20110112698A1 (en) | Method and apparatus for energy and emission reduction | |
| CN113240211B (en) | Method and device for predicting wastewater discharge, computer equipment and storage medium | |
| CN115186983A (en) | An automatic carbon emission accounting system based on online monitoring of energy consumption | |
| CN111523794B (en) | Research and judgment method of environmental protection management and control measures based on electricity consumption characteristics of polluting enterprises | |
| CN119740888B (en) | Carbon emission management system and method based on digital twinning | |
| CN113256037B (en) | Unsteady state wastewater discharge prediction method, unsteady state wastewater discharge prediction device, computer equipment and storage medium | |
| CN118261467A (en) | Calculation method for emission reduction implementation effect analysis of park enterprises in heavily polluted weather | |
| CN113807743A (en) | Power grid dispatching automation software reliability assessment method and system | |
| CN119831455A (en) | Regional carbon emission reduction potential analysis method based on electric power big data | |
| CN120087756A (en) | Pollutant discharge permit evaluation and management method based on intelligent analysis | |
| CN116128277A (en) | Dynamic early warning method and system for classification and grading of fixed pollution sources | |
| CN118229157B (en) | Hydropower efficiency monitoring method, device, computer equipment, readable storage medium and program product | |
| CN117371795A (en) | Supervision system based on environment comprehensive big data | |
| CN116263850B (en) | Online sewage water quality early warning method combining offline simulation data | |
| CN114089257B (en) | Electric energy meter burning on-line monitoring method, system and medium | |
| CN118863901A (en) | A method for tracing the root causes of pollutants emitted from steel production based on deep learning | |
| CN114116843B (en) | Operation monitoring method of small-scale sewage treatment facilities in villages and towns based on electricity consumption data | |
| CN117391554A (en) | Energy-carbon accounting method based on energy consumption data | |
| CN115374092A (en) | A dynamic time window sewage treatment monitoring data rationality judgment method | |
| CN116865246A (en) | A method and system for predicting the feasible region of industrial user load based on fast response | |
| CN113723775A (en) | Enterprise and industry operation risk assessment method based on electric power big data | |
| Sha et al. | Cost Risk Assessment of Power Transmission and Transformation Project Based on MCS-GA-ELM | |
| CN118888028A (en) | A quantitative method and system for reducing emissions of atmospheric pollutants from industrial sources based on a quantitative model |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |