Industrial and commercial energy storage peak shaving method based on artificial intelligence

Abstract

The invention discloses an industrial and commercial energy storage peak regulation method based on artificial intelligence, and relates to the technical field of industrial and commercial energy storage peak regulation, and the method comprises the following steps: in the discharge process of an energy storage system, continuously collecting the theoretical discharge capacity and the actual discharge capacity of the energy storage system, obtaining the discharge data of the energy storage system, and collecting the environment temperature data; performing abnormal analysis processing to obtain standard energy storage discharge data and standard temperature data; efficiency calculation and analysis are carried out according to the standard energy storage discharge data, and a discharge efficiency prediction model is constructed according to the standard temperature data; predicting the discharge efficiency based on the discharge efficiency prediction model, and adjusting the target storage electric quantity of the energy storage system; the method is used for solving the problem that when an existing industrial and commercial energy storage peak regulation technology configures the electric quantity stored in an energy storage system, the discharge efficiency of the future date cannot be predicted according to the discharge efficiency data of the energy storage system and the environment temperature data, and the accuracy of electric quantity configuration cannot be guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial and commercial energy storage peak shaving, in particular to an industrial and commercial energy storage peak shaving method based on artificial intelligence. BACKGROUND

[0002] Industrial and commercial energy storage peak shaving technology refers to a user-side electric energy optimization management technology that faces industrial and commercial users, takes the core target of stabilizing the fluctuation of power consumption, reducing the cost of peak power consumption, and reducing the impact on the public power grid, and realizes the storage of low-valley period and the discharge of peak period through the charge and discharge time sequence control of the energy storage system. It is the core technology for industrial and commercial users to improve the economy and stability of power consumption.

[0003] The existing industrial and commercial energy storage peak shaving technology often establishes a prediction model to predict the power consumption of future dates when configuring the power stored in the energy storage system. The predicted power is corrected with a fixed discharge efficiency to ensure that the power consumption demand of future dates is met. The influence of environmental temperature on discharge efficiency is often ignored. In fact, environmental temperature has a direct impact on the discharge efficiency of the energy storage system. As the environmental temperature rises, the internal heat dissipation pressure of the battery increases, which will cause the actual discharge efficiency to be low. The decrease of environmental temperature will reduce the discharge efficiency through the dual effects of inhibiting electrochemical reaction and increasing internal resistance loss. Ignoring the influence of environmental temperature changes on discharge efficiency in future dates will bring many problems to energy storage peak shaving. The energy storage power configuration scheme calculated according to the fixed discharge efficiency will cause the discharge efficiency to fluctuate due to the actual temperature change, which is easy to cause the mismatch between the actual output power of the energy storage system and the power consumption demand. It may cause a power supply gap during the peak power consumption period, and the emergency power supply mechanism has to be started, which will increase the power supply cost. The originally planned power storage and release rhythm is disturbed, and part of the power to be allocated cannot be integrated into the peak shaving process as expected, which reduces the actual utilization efficiency of the energy storage system and makes it difficult to achieve the designed effect. Therefore, the existing industrial and commercial energy storage peak shaving technology cannot predict the discharge efficiency of future dates according to the discharge efficiency data and environmental temperature data of the energy storage system when configuring the power stored in the energy storage system, so as to ensure the accuracy of power configuration. SUMMARY

[0004] The present application aims to at least solve one of the technical problems in the prior art, by continuously collecting the theoretical discharge power and the actual discharge power of the energy storage system during the discharge process of the energy storage system, obtaining energy storage system discharge data, and collecting environmental temperature data; performing abnormal analysis and processing on the energy storage system discharge data and the environmental temperature data respectively to obtain standard energy storage discharge data and standard temperature data; performing efficiency calculation analysis according to the standard energy storage discharge data, and constructing a discharge efficiency prediction model according to the standard temperature data; predicting the discharge efficiency based on the discharge efficiency prediction model, and adjusting the target storage power of the energy storage system; to solve the problem that the existing industrial and commercial energy storage peak shaving technology cannot predict the discharge efficiency of future dates according to the discharge efficiency data and the environmental temperature data of the energy storage system when configuring the power stored by the energy storage system, and cannot ensure the accuracy of power configuration.

[0005] To achieve the above-mentioned purpose, the present application provides an industrial and commercial energy storage peak shaving method based on artificial intelligence, comprising the following steps: During the discharge process of the energy storage system, the theoretical discharge power and the actual discharge power of the energy storage system are continuously collected to obtain energy storage system discharge data, and environmental temperature data is collected; The energy storage system discharge data and the environmental temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharge data and standard temperature data; Efficiency calculation analysis is performed according to the standard energy storage discharge data, and a discharge efficiency prediction model is constructed according to the standard temperature data; The discharge efficiency is predicted based on the discharge efficiency prediction model, and the target storage power of the energy storage system is adjusted.

[0006] Further, during the discharge process of the energy storage system, the theoretical discharge power and the actual discharge power of the energy storage system are continuously collected to obtain energy storage system discharge data, and environmental temperature data is collected, comprising the following sub-steps: Any one energy storage system is denoted as a first energy storage system, during the discharge process of the first energy storage system, all loads connected to the first energy storage system are denoted as energy storage discharge loads, the power reduced by the first energy storage system is denoted as the theoretical discharge power of the first energy storage system, and the sum of the power actually obtained by all energy storage discharge loads is denoted as the actual discharge power of the first energy storage system; And obtain the daily discharge period of the first energy storage system, denoted as the energy storage discharge period; any date of discharge of the first energy storage system is denoted as the first date.

[0007] Further, during the discharge process of the energy storage system, the theoretical discharge power and the actual discharge power of the energy storage system are continuously collected to obtain energy storage system discharge data, and environmental temperature data is collected, comprising the following sub-steps: The first period size is set as t1, and for the first date, the theoretical discharge capacity and the actual discharge capacity of the first energy storage system in each first period are periodically collected during the discharging process of the first energy storage system, and the collection time is recorded, which are arranged in time sequence respectively and recorded as the theoretical discharge sequence and the actual discharge sequence of the first date, respectively, and marked as the energy storage system discharge data of the first date; Meanwhile, the ambient temperature of the external environment where the first energy storage system is located is collected at a first time interval during the discharging process of the first energy storage system, and the collection time is recorded, which is arranged in time sequence and recorded as the ambient temperature sequence of the first date, and marked as the ambient temperature data; The energy storage system discharge data and the ambient temperature data of each date are repeatedly collected.

[0008] Further, the energy storage system discharge data and the ambient temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharge data and standard temperature data, including the following sub-steps: According to the theoretical discharge sequence and the actual discharge sequence of the first date, the actual discharge capacity of each period is divided by the theoretical discharge capacity to obtain the discharge efficiency, and the discharge efficiency sequence is obtained after completion; The coefficient of variation of the discharge efficiency sequence is calculated and recorded as CV0; any discharge efficiency in the discharge efficiency sequence is recorded as the first discharge efficiency PN(i), where i represents the position number, and the initial window size is set as k1; The initial window with the center position of PN(i) is recorded as the initial neighborhood window of PN(i); the coefficient of variation of the data in the initial neighborhood window of PN(i) is calculated and recorded as CV1; If CV1 is not greater than CV0, the initial neighborhood window of PN(i) is recorded as the reference window i of PN(i); if CV1 is greater than CV0, the center position of the initial window is kept unchanged, the size of the initial window is expanded to k2*k1, and the expanded initial window is recorded as the reference window i of PN(i), where k2 is a set proportion coefficient.

[0009] Further, the energy storage system discharge data and the ambient temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharge data and standard temperature data, including the following sub-steps: For the reference window i, the relative difference XP(i) corresponding to PN(i) is calculated, where XP(i)=|PN(i)-PN(i-1)| / PN(i-1); the relative differences corresponding to all data in the reference window i are repeatedly calculated to obtain the relative difference set i, and the mean and the standard deviation of the relative difference set i are calculated and recorded in sequence as AP1 and AB1, respectively; If XP(i) is not located in [AP1-k3*AB1, AP1+k3*AB1], mark PN(i) as a trend abnormality suspected point; otherwise, do not mark, and repeat marking all discharge efficiencies in the discharge efficiency sequence, wherein k3 is a set proportion coefficient; Calculate the mean and standard deviation of all data in the reference window i, and mark them as AP2 and AB2 in order, respectively, and if PN(i) is not located in [AP1-k4*AB1, AP1+k4*AB1], mark PN(i) as a fluctuation abnormality candidate point, wherein k4 is a set proportion coefficient; If PN(i) is a fluctuation abnormality candidate point, divide all data in the reference window i by using a density clustering algorithm to obtain multiple clusters, and mark the cluster with the largest data quantity as a normal cluster, and if PN(i) is not located in the normal cluster, mark PN(i) as a fluctuation abnormality suspected point, otherwise, do not mark; If PN(i) is a fluctuation abnormality suspected point and PN(i) is a trend abnormality suspected point, mark PN(i) as abnormal data, otherwise, mark PN(i) as normal data; repeat marking all discharge efficiencies in the discharge efficiency sequence; and after completion, obtain the marked efficiency sequence of the first date, and mark it as the standard energy storage discharge data of the first date.

[0010] Further, the abnormality analysis and processing of the energy storage system discharge data and the environmental temperature data respectively to obtain the standard energy storage discharge data and the standard temperature data further comprises the following sub-steps: For the environmental temperature sequence of the first date, mark any environmental temperature in the environmental temperature sequence as HT(j), wherein j represents the position sequence number; if HT(j) is greater than two adjacent environmental temperatures or less than two adjacent environmental temperatures at the same time, mark HT(j) as a split point; repeat obtaining all split points, and divide the environmental temperature sequence into multiple subsequences according to all split points; and mark the subsequence where HT(j) is located as a first subsequence; Perform nonlinear fitting on the first subsequence to obtain a corresponding fitting function, and mark it as a first fitting function; and obtain a fitting value corresponding to HT(j) according to the first fitting function and the position corresponding to HT(j); If k5 environmental temperatures before and after HT(j) are taken as neighborhood reference temperatures, and the corresponding fitting values are obtained, wherein k5 is a set number; If HT(j) is greater than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures less than the corresponding fitting value, mark HT(j) as abnormal data; if HT(j) is less than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures greater than the corresponding fitting value, mark HT(j) as abnormal data; otherwise, mark HT(j) as normal data, wherein k6 is a set proportion coefficient; Repeat the marking for all ambient temperatures in the ambient temperature sequence, and after completion, the corresponding marked temperature sequence is obtained, denoted as the standard temperature data of the first date.

[0011] Further, the efficiency calculation analysis is performed according to the standard energy storage discharge data, and the discharge efficiency prediction model is constructed according to the standard temperature data, including the following sub-steps: Set the first time length as t2, and divide the marked efficiency sequence and the marked temperature sequence of the first date into multiple corresponding segments according to t2, respectively denoted as the divided efficiency sequence and the divided temperature sequence; Denote any segment with a time length of t2 in the divided efficiency sequence as a first efficiency segment, denote any discharge efficiency in the first efficiency segment as a first efficiency PG, and set the first window size as k7; Take PG as the starting position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the first reference deviation GB1 of PG; and take PG as the middle position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the second reference deviation GB2 of PG; and take PG as the end position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the third reference deviation GB3 of PG; Calculate the weight deviation QG of PG, where QG=1 / [(GB1+GB2+GB3) / 3]; repeat the weight deviation of all discharge efficiencies in the first efficiency segment, and sum them up, denoted as the weight reference QW0; Calculate the weighted efficiency QP corresponding to PG, where QP=(QG / QW0)*PG, repeat the weighted efficiency corresponding to all discharge efficiencies in the first efficiency segment, and sum them up to obtain the reference discharge efficiency of the first efficiency segment; Repeat the reference discharge efficiency in the divided efficiency sequence, and arrange them in the corresponding order, denoted as the reference discharge efficiency sequence of the first date.

[0012] Further, the efficiency calculation analysis is performed according to the standard energy storage discharge data, and the discharge efficiency prediction model is constructed according to the standard temperature data, including the following sub-steps: For the divided temperature sequence, denote any segment with a time length of t2 in the divided temperature sequence as a first temperature segment, perform nonlinear fitting on the first temperature segment to obtain the corresponding fitting function, and then calculate the average temperature of the first temperature segment according to the corresponding fitting function, repeat the average temperature in the divided temperature sequence, and arrange them in the corresponding order, denoted as the average temperature sequence of the first date; Denote the reference discharge efficiency sequence and the average temperature sequence of the first date as the temperature efficiency data of the first date; repeat the temperature efficiency data of each day, denoted as the temperature efficiency reference data.

[0013] Further, the efficiency calculation analysis according to the standard energy storage discharge data and the discharge efficiency prediction model constructed according to the standard temperature data further include the following sub-steps: The original prediction model is constructed based on the LSTM model, the original prediction model includes an input layer, a hidden layer and an output layer, the output of the original prediction model is set as the reference discharge efficiency sequence of the future date; the original prediction model is trained by using the temperature efficiency reference data, and the discharge efficiency prediction model is obtained after the training.

[0014] Further, the discharge efficiency is predicted based on the discharge efficiency prediction model, and the target storage power of the energy storage system is adjusted, including the following sub-steps: The average temperature sequence corresponding to the future date is obtained, the discharge efficiency of the future date is predicted by using the discharge efficiency prediction model, and the reference discharge efficiency sequence of the future date is obtained; The average discharge efficiency of the future date is calculated according to the reference discharge efficiency sequence of the future date; the estimated storage power of the energy storage system for the future date is obtained, which is recorded as the target storage power of the energy storage system; The target storage power of the energy storage system is corrected by using the average discharge efficiency of the future date, and the corrected storage power of the energy storage system for the future date is obtained, and the energy storage is performed according to the corrected storage power.

[0015] The present application has the following advantages: in the process of discharging the energy storage system, the theoretical discharge power and the actual discharge power of the energy storage system are continuously collected to obtain the energy storage system discharge data, and the environmental temperature data is collected; the energy storage system discharge data and the environmental temperature data are respectively analyzed and processed to obtain standard energy storage discharge data and standard temperature data; the efficiency calculation analysis is performed according to the standard energy storage discharge data, and the discharge efficiency prediction model is constructed according to the standard temperature data; the discharge efficiency is predicted based on the discharge efficiency prediction model, and the target storage power of the energy storage system is adjusted; when the power stored in the energy storage system is configured, the prediction model can be established according to the discharge efficiency data of the energy storage system and the environmental temperature data to predict the discharge efficiency of the future date, and the accuracy of the power configuration is ensured. The application judges whether to expand the window based on the coefficient of variation, so that the anomaly detection can adaptively adjust the reference window size according to the data fluctuation around the point; in this way, the calculation amount can be reduced in the stable section, and the field of view can be expanded in the fluctuation section to obtain more robust judgment and reduce the misjudgment rate; the relative difference is used to detect the trend direction of the mutation, and the density clustering is used to detect the fluctuation anomaly suspected point, which can distinguish the mutation type anomaly from the continuous jitter type anomaly, and is more accurate than the single threshold method; the standard deviation of three position windows is calculated for each discharge efficiency, the weight standard deviation is obtained, and then the weight standard deviation is weighted, so that the data points that are more stable in the time period can be given greater weight, so that the reference discharge efficiency of the section is mainly composed of stable and reliable data, and the accidental jitter is avoided to ensure that the subsequent model can learn the correct feature relationship. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A step flowchart of the method of the application; Figure 2 A trend anomaly suspected point marking flowchart of the application; Figure 3 A reference discharge efficiency calculation flowchart of the application; Figure 4 A structural schematic diagram of an electronic device of the application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0018] Embodiment 1, please refer to Figure 1 As shown in the drawings, the application provides an industrial and commercial energy storage peak shaving method based on artificial intelligence, which comprises the following steps: Step S1, during the discharge process of the energy storage system, continuously collecting the theoretical discharge capacity and the actual discharge capacity of the energy storage system to obtain energy storage system discharge data, and collecting environmental temperature data; step S1 comprises the following substeps: Step S101, taking any one energy storage system as a first energy storage system, during the discharge process of the first energy storage system, taking all loads connected with the first energy storage system as energy storage discharge loads, taking the reduced amount of the first energy storage system as the theoretical discharge capacity of the first energy storage system, that is, the energy reduction amount seen from the energy storage system side, representing the energy output by the system; and taking the sum of the actual amounts of electricity obtained by all energy storage discharge loads as the actual discharge capacity of the first energy storage system; that is, the actual received energy on the load side, representing the energy actually transmitted to the load; Step S102, and obtain the daily discharge period of each first energy storage system, denoted as the energy storage discharge period; the date of discharge of any first energy storage system is denoted as the first date; the energy storage system does not necessarily discharge all day; obtaining the discharge period only samples and analyzes the actual discharge period, avoiding mixing idle or charging time into the discharge efficiency calculation, thereby improving the analysis pertinence.

[0019] Step S103, set the first period size as t1, for the first date, periodically collect the theoretical discharge capacity and the actual discharge capacity of the first energy storage system in each first period during the discharge of the first energy storage system, and record the collection time, which are arranged in time sequence respectively, denoted as the theoretical discharge sequence and the actual discharge sequence of the first date, and marked as the energy storage system discharge data of the first date; in this example, t1 = 1 second, which can be flexibly set; Step S104, simultaneously in the process of discharging the first energy storage system, collect the ambient temperature of the external environment where the first energy storage system is located at the first time interval, and record the collection time, which are arranged in time sequence, denoted as the ambient temperature sequence of the first date, and marked as the ambient temperature data; if the first energy storage system is in the room, the indoor ambient temperature needs to be collected; the ambient temperature of the future date used for subsequent prediction can be obtained through weather forecast, and the indoor ambient temperature of the future date needs to be predicted by oneself to meet the subsequent needs; Step S105, repeat the collection of the energy storage system discharge data and the ambient temperature data of each date; In the specific implementation process, for the mainstream lithium battery energy storage system, the decrease of the ambient temperature will reduce the discharge efficiency through the dual effects of inhibiting the electrochemical reaction and increasing the internal resistance loss; the increase of the ambient temperature; in the short term, it will reduce the viscosity of the electrolyte and speed up the ion migration, and the discharge efficiency may be slightly improved, but the increase of the ambient temperature will increase the heat dissipation pressure inside the battery, and in the case of long-time discharge, heat accumulation is easy to occur, which will cause the energy loss to increase, which will cause the discharge efficiency to be lower than the conventional level; therefore, the ambient temperature has a direct impact on the discharge efficiency of the energy storage system.

[0020] Step S2, the energy storage system discharge data and the ambient temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharge data and standard temperature data; step S2 includes the following substeps: Step S201, according to the theoretical discharge sequence and the actual discharge sequence of the first date, the actual discharge capacity of each period is divided by the theoretical discharge capacity, denoted as the discharge efficiency, and the discharge efficiency sequence is obtained after completion; Step S202, please refer to Figure 2As shown, the coefficient of variation of the discharge efficiency sequence is calculated, denoted as CV0; any one of the discharge efficiency in the discharge efficiency sequence is denoted as the first discharge efficiency PN(i), wherein i represents the position number, and the initial window size is set as k1; in this embodiment, k1=20, which can be flexibly set according to the actual application scenario; CV reflects the relative fluctuation intensity of the data in the window; the greater CV is, the more intense the fluctuation is; In step S203, the initial window is denoted as the initial neighborhood window of PN(i) with PN(i) as the center position of the initial window; the coefficient of variation of the data in the initial neighborhood window of PN(i) is calculated, denoted as CV1; In step S204, if CV1 is not greater than CV0, the initial neighborhood window of PN(i) is denoted as the reference window i of PN(i); if CV1 is greater than CV0, the center position of the initial window is kept unchanged, and the size of the initial window is expanded to k2*k1, and the expanded initial window is denoted as the reference window i of PN(i), wherein k2 is a set proportion coefficient; k2>1, in this embodiment, k2=2, which can be flexibly set; the actual discharge efficiency sequence often has a local stationary section and a local fluctuation section; the fixed small window may frequently misjudge in the fluctuation area due to insufficient data, and the fixed large window may also mask the short-time mutation; therefore, adaptively comparing the local coefficient of variation with the global coefficient of variation can dynamically determine whether the small window or the large window is used; when the data is already fluctuated greatly, expanding the window can reduce the probability of marking the normal high fluctuation as abnormal.

[0021] In step S205, for the reference window i, the relative difference XP(i) corresponding to PN(i) is calculated, wherein XP(i)=|PN(i)-PN(i-1)| / PN(i-1); the relative differences corresponding to all the data in the reference window i are repeatedly calculated to obtain the relative difference set i, and the mean and the standard deviation of the relative difference set i are calculated, which are sequentially denoted as AP1 and AB1 respectively; the relative difference directly quantifies the instantaneous slope or the mutation amplitude, and the relative difference can more accurately reflect the significance of the change than the absolute difference, and can accurately identify the trend anomaly; In step S206, if XP(i) is not located in [AP1-k3*AB1, AP1+k3*AB1], PN(i) is marked as a trend anomaly suspected point; otherwise, it is not marked, and the marking is repeated for all the discharge efficiencies in the discharge efficiency sequence, wherein k3 is a set proportion coefficient; in this embodiment, k3=3, which can be flexibly set; Step S207: Calculate the mean and standard deviation of all data in reference window i, and denot them as AP2 and AB2 respectively. If PN(i) is not located in [AP1-k4*AB1, AP1+k4*AB1], then mark PN(i) as a candidate point for fluctuation anomaly, where k4 is the set proportional coefficient; in this embodiment, k4=2, which can be set flexibly; simply using the mean and threshold is not accurate enough and may misjudge real anomalies and normal fluctuations, so further judgment is needed on the candidate points for fluctuation anomalies. Step S208: If PN(i) is a candidate point for fluctuation anomaly, then the density clustering algorithm is used to divide all data in the reference window i into multiple clusters, and the cluster with the largest amount of data is recorded as the normal cluster. If PN(i) is not located in the normal cluster, then PN(i) is marked as a suspected point for fluctuation anomaly; otherwise, it is not marked. The core feature of a real anomaly point, such as acquisition error or sensor failure, is that it exists in isolation in the local time series and has no similar efficiency value around it. However, normal changes caused by load fluctuation will form local clusters, and the surrounding efficiency values ​​change synchronously with the load, resulting in high density. Therefore, if PN(i) is not located in the normal cluster, then PN(i) is marked as a suspected point for fluctuation anomaly. Step S209: If PN(i) is a suspected point of fluctuation abnormality and PN(i) is a suspected point of trend abnormality, then mark PN(i) as abnormal data; otherwise, mark PN(i) as normal data. Repeat the marking of all discharge efficiencies in the discharge efficiency sequence. After completion, the marked efficiency sequence of the first date is obtained and recorded as the standard energy storage discharge data of the first date. It is required that both types of abnormal conditions be met at the same time before it is finally judged as abnormal, which helps to reduce the misjudgment of real but rare discharge efficiency values.

[0022] Step S210: For the ambient temperature sequence of the first date, any ambient temperature in the ambient temperature sequence is denoted as HT(j), where j represents the position number; if HT(j) is simultaneously greater than or simultaneously less than two adjacent ambient temperatures, then HT(j) is marked as a split point; all split points are repeatedly obtained, and the ambient temperature sequence is divided into multiple subsequences according to all split points; the subsequence where HT(j) is located is denoted as the first subsequence; local extrema are detected as split points to cut the temperature sequence into several segments, making each segment closer to a single trend and improving the accuracy of fitting; Step S211: Perform nonlinear fitting on the first subsequence to obtain the corresponding fitting function, denoted as the first fitting function. Obtain the fitting value corresponding to HT(j) based on the first fitting function and the position corresponding to HT(j). Step S212: If k5 ambient temperatures before and after HT(j) are taken as neighborhood reference temperatures, and the corresponding fitted values ​​are obtained, where k5 is the number of values ​​set; in this embodiment, k5=2, generally 2 or 3; Step S213, if HT(j) is greater than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures less than the corresponding fitting value, mark HT(j) as abnormal data; if HT(j) is less than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures greater than the corresponding fitting value, mark HT(j) as abnormal data; otherwise, mark HT(j) as normal data, wherein k6 is a set proportion coefficient; in this embodiment, k6 = 0.9, which can be flexibly set; the nonlinear fitting gives the theoretical or expected temperature of the point in the current segment. In the natural environment, normal fluctuations in temperature will cause continuous multiple points to deviate slightly from the trend, for example, wind gusts will cause a short-term temperature drop, and the subsequent trend will recover. However, collection errors are usually isolated deviations, that is, the deviation from the field point is not the same. Comparing the actual measurement with the expectation in combination with the neighborhood distribution can distinguish between transient disturbances and real abnormal readings; Step S214, repeat the marking of all ambient temperatures in the ambient temperature sequence, and complete the corresponding marked temperature sequence, denoted as the standard temperature data of the first date; In the specific implementation process, the theoretical discharge sequence and the actual discharge sequence are affected by the load connected to the energy storage system and may suddenly change. It is not possible to distinguish between normal fluctuations and false mutations. However, the discharge efficiency does not change suddenly and is relatively stable. Therefore, calculating the corresponding discharge efficiency sequence and analyzing and processing the discharge efficiency sequence can reduce the analysis difficulty and reduce the calculation amount.

[0023] Step S3, efficiency calculation and analysis are performed according to the standard energy storage discharge data, and a discharge efficiency prediction model is constructed according to the standard temperature data; step S3 includes the following substeps: Step S301, set the first time length as t2, and divide the marked efficiency sequence and the marked temperature sequence of the first date into multiple corresponding segments according to t2, respectively denoted as the divided efficiency sequence and the divided temperature sequence; in this embodiment, t2 = 15 minutes, which is also the minimum time scale for subsequent prediction; the sampling frequency of the original data is very high, and direct prediction is costly and difficult, and is also easy to be dominated by noise; segmentation can change the data into stable statistics, reducing the prediction difficulty; Step S302, please refer to Figure 3 The divided efficiency sequence is denoted as a first efficiency segment, and any discharge efficiency in the first efficiency segment is denoted as a first efficiency PG. The first window size is set as k7; in this embodiment, k7 = 10, which can be flexibly set; Step S303, taking PG as the start position of the first window, and calculating the standard deviation of the charging efficiency in the first window at this time, denoted as the first reference standard deviation GB1 of PG; and taking PG as the middle position of the first window, and calculating the standard deviation of the charging efficiency in the first window at this time, denoted as the second reference standard deviation GB2 of PG; and taking PG as the end position of the first window, and calculating the standard deviation of the charging efficiency in the first window at this time, denoted as the third reference standard deviation GB3 of PG; taking the average of the three positions can more stably reflect the local stability of the point at different relative positions.

[0024] Step S304, calculating the weight standard deviation QG of PG, wherein QG = 1 / [(GB1+GB2+GB3) / 3]; repeatedly obtaining the weight standard deviation of all discharge efficiencies in the first efficiency segment, and summing, denoted as the weight reference QW0; Step S305, calculating the weighted efficiency QP corresponding to PG, wherein QP = (QG / QW0)*PG, repeatedly obtaining the weighted efficiency corresponding to all discharge efficiencies in the first efficiency segment, and summing to obtain the reference discharge efficiency of the first efficiency segment; directly using the average of the discharge efficiencies in the segment will pull the results of the fluctuation points; by weighting through the standard deviation, the data points that are more stable in the time period can be given greater weight, so that the reference discharge efficiency of the segment is mainly composed of stable data points, rather than being pulled by occasional fluctuations, and the reference discharge efficiency is more biased towards the long-term stable contributors in the segment, thereby improving the robustness of the segment reference value; Step S306, repeatedly obtaining all reference discharge efficiencies in the divided efficiency sequence, and arranging them in the corresponding order, denoted as the reference discharge efficiency sequence of the first date.

[0025] Step S307, for the divided temperature sequence, taking any segment with a time length of t2 in the divided temperature sequence as the first temperature segment, performing nonlinear fitting on the first temperature segment to obtain the corresponding fitting function, and then calculating the average temperature of the first temperature segment according to the corresponding fitting function, repeatedly obtaining all average temperatures in the divided temperature sequence, and arranging them in the corresponding order, denoted as the average temperature sequence of the first date; the first temperature segment may contain noise; using the fitting function to average the segment can better reflect the intrinsic temperature level or trend of the segment, reducing the influence of noise; for example, the temperature may show a slow rise or fall within the segment; the fitting curve can capture this trend and make the average temperature of the segment more physically meaningful; Step S308, taking the reference discharge efficiency sequence and the average temperature sequence of the first date as the temperature efficiency data of the first date; repeatedly obtaining the temperature efficiency data of each day, denoted as the temperature efficiency reference data.

[0026] Step S309, an original prediction model is constructed based on the LSTM model, the original prediction model including an input layer, a hidden layer and an output layer, the original prediction model being set to output a reference discharge efficiency sequence of a future date; the original prediction model is trained by using temperature efficiency reference data, and a discharge efficiency prediction model is obtained after the training; the relationship between temperature and discharge efficiency is often nonlinear, and the LSTM can fit the complex nonlinear mapping; In the specific implementation process, if the PG is at the edge of the segment, the first window cannot meet all placement requirements, and there is only one or two reference standard deviations, then the average value is calculated directly by the obtained standard deviation, and subsequent processing is performed, without the need to reach 3.

[0027] Step S4, the discharge efficiency is predicted based on the discharge efficiency prediction model, and the target storage capacity of the energy storage system is adjusted; step S4 includes the following sub-steps: Step S401, an average temperature sequence corresponding to a future date is obtained, the discharge efficiency of the future date is predicted by using the discharge efficiency prediction model, and a reference discharge efficiency sequence of the future date is obtained; Step S402, the average discharge efficiency of the future date is calculated according to the reference discharge efficiency sequence of the future date; the estimated storage capacity of the energy storage system for the future date is obtained, denoted as the target storage capacity of the energy storage system, that is, the electricity consumption of the future date predicted by using an algorithm or a model; Step S403, the target storage capacity of the energy storage system is corrected by using the average discharge efficiency of the future date, to obtain a corrected storage capacity of the energy storage system for the future date, and the energy storage is performed according to the corrected storage capacity; for example, the target storage capacity is 100 kWh, and the average discharge efficiency predicted for tomorrow is 90%, so at least (100 / 0.9) kWh of electricity needs to be stored to ensure the basic electricity demand of tomorrow; In the specific implementation process, the average discharge efficiency of the future date can also be weighted and averaged according to the importance or possible electricity consumption of each period, and the specific correction method of the target storage capacity can also be set and adjusted according to the actual application scenario.

[0028] Embodiment 2, please refer to Figure 4 as shown, Figure 4An example is provided for a structural diagram of an electronic device, which can include a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete the communication among each other through the communication bus. The memory stores computer readable instructions, and the processor can invoke the instructions in the memory, and when the computer readable instructions are executed by the processor, the steps in the artificial intelligence-based industrial and commercial energy storage peak shaving method are run to realize the following functions: during the discharging process of the energy storage system, continuously collecting the theoretical discharge power and the actual discharge power of the energy storage system to obtain energy storage system discharge data, and collecting environmental temperature data; performing abnormal analysis and processing on the energy storage system discharge data and the environmental temperature data respectively to obtain standard energy storage discharge data and standard temperature data; performing efficiency calculation and analysis according to the standard energy storage discharge data, and constructing a discharge efficiency prediction model according to the standard temperature data; predicting the discharge efficiency based on the discharge efficiency prediction model, and adjusting the target storage power of the energy storage system.

[0029] In addition, the logic instructions in the memory described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0030] In embodiment 3, the present application also provides a computer readable storage medium, and the present application provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to run the steps in the artificial intelligence-based industrial and commercial energy storage peak shaving method to realize the following functions: during the discharging process of the energy storage system, continuously collecting the theoretical discharge power and the actual discharge power of the energy storage system to obtain energy storage system discharge data, and collecting environmental temperature data; performing abnormal analysis and processing on the energy storage system discharge data and the environmental temperature data respectively to obtain standard energy storage discharge data and standard temperature data; performing efficiency calculation and analysis according to the standard energy storage discharge data, and constructing a discharge efficiency prediction model according to the standard temperature data; predicting the discharge efficiency based on the discharge efficiency prediction model, and adjusting the target storage power of the energy storage system.

[0031] Through the description of the above embodiments, the embodiments of the present application can be provided as a method, a system or a computer program product. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

[0032] In the embodiments provided by the present application, it should be understood that the disclosed system or method can be implemented in other manners. The embodiments described above are merely schematic, and should not be construed as limiting. For example, the division of the modules or the units is merely logical function division, and there can be other division manners in actual implementation. For example, a plurality of modules or units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different modules can be indirect couplings or communication connections through some interfaces, and there can be electric, mechanical or other forms.

[0033] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit them; even if the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An artificial intelligence based industrial and commercial energy storage peak shaving method, characterized in that, The method comprises the following steps: During the discharging process of the energy storage system, the theoretical discharge capacity and the actual discharge capacity of the energy storage system are continuously collected to obtain the energy storage system discharging data, and the ambient temperature data are collected; The energy storage system discharging data and the ambient temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharging data and standard temperature data; Efficiency calculation and analysis are performed according to the standard energy storage discharging data, and a discharging efficiency prediction model is constructed according to the standard temperature data; The discharging efficiency is predicted based on the discharging efficiency prediction model, and the target storage capacity of the energy storage system is adjusted.

2. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 1, characterized in that, During the discharging process of the energy storage system, the theoretical discharge capacity and the actual discharge capacity of the energy storage system are continuously collected to obtain the energy storage system discharging data, and the ambient temperature data are collected, which comprises the following sub-steps: Any one energy storage system is recorded as a first energy storage system, during the discharging process of the first energy storage system, all loads connected to the first energy storage system are recorded as energy storage discharging loads, the reduced capacity of the first energy storage system is recorded as the theoretical discharge capacity of the first energy storage system, and the sum of the actual capacities obtained by all energy storage discharging loads is recorded as the actual discharge capacity of the first energy storage system; The discharging time period of the first energy storage system is obtained and recorded as the energy storage discharging time period; and the discharging date of any one first energy storage system is recorded as a first date.

3. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 2, characterized in that, During the discharging process of the energy storage system, the theoretical discharge capacity and the actual discharge capacity of the energy storage system are continuously collected to obtain the energy storage system discharging data, and the ambient temperature data are collected, which further comprises the following sub-steps: The first cycle size is set as t1, for the first date, during the discharging process of the first energy storage system, the theoretical discharge capacity and the actual discharge capacity of the first energy storage system in each first cycle are periodically collected, and the collection time is recorded, which are arranged in time sequence respectively, and recorded as the theoretical discharging sequence and the actual discharging sequence of the first date respectively, and marked as the energy storage system discharging data of the first date; Meanwhile, during the discharging process of the first energy storage system, the ambient temperature of the external environment where the first energy storage system is located is collected at a first time interval, and the collection time is recorded, which is arranged in time sequence and recorded as the ambient temperature sequence of the first date, and marked as the ambient temperature data; The energy storage system discharging data and the ambient temperature data of each date are repeatedly collected.

4. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 3, characterized in that, The energy storage system discharging data and the ambient temperature data are respectively subjected to abnormal analysis and processing to obtain standard energy storage discharging data and standard temperature data, which comprises the following sub-steps: According to the theoretical discharging sequence and the actual discharging sequence of the first date, the actual discharge capacity of each cycle is divided by the theoretical discharge capacity to obtain the discharging efficiency, and the discharging efficiency sequence is obtained after completion; The coefficient of variation of the discharging efficiency sequence is calculated and recorded as CV0; any one discharging efficiency in the discharging efficiency sequence is recorded as a first discharging efficiency PN(i), wherein i represents the position number, and the initial window size is set as k1; The initial window of this time is recorded as the initial neighborhood window of PN(i) with PN(i) as the center position of the initial window; the coefficient of variation of the data in the initial neighborhood window of PN(i) is calculated and recorded as CV1; If CV1 is not greater than CV0, the initial neighborhood window of PN(i) is recorded as the reference window i of PN(i), if CV1 is greater than CV0, the center position of the initial window is kept unchanged, the size of the initial window is expanded to k2*k1, and the expanded initial window is recorded as the reference window i of PN(i), wherein k2 is a set proportion coefficient.

5. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 4, characterized in that, The abnormal analysis and processing are performed on the energy storage system discharge data and the environmental temperature data respectively to obtain standard energy storage discharge data and standard temperature data, and the following sub-steps are further included. For the reference window i, a relative difference XP(i) corresponding to PN(i) is calculated, wherein XP(i)=|PN(i)-PN(i-1)| / PN(i-1); the relative differences corresponding to all data in the reference window i are repeatedly calculated to obtain a relative difference set i, and the mean value and the standard deviation of the relative difference set i are calculated and recorded as AP1 and AB1 in sequence respectively; If XP(i) is not located in [AP1-k3*AB1, AP1+k3*AB1], PN(i) is marked as a trend abnormal suspected point; otherwise, PN(i) is not marked, and the marking is repeated for all discharge efficiencies in the discharge efficiency sequence, wherein k3 is a set proportion coefficient; The mean value and the standard deviation of all data in the reference window i are calculated and recorded as AP2 and AB2 in sequence respectively, and if PN(i) is not located in [AP1-k4*AB1, AP1+k4*AB1], PN(i) is marked as a fluctuation abnormal candidate point, wherein k4 is a set proportion coefficient; If PN(i) is a fluctuation abnormal candidate point, density clustering algorithm is used to divide all data in the reference window i to obtain multiple clusters, and the cluster with the largest data quantity is recorded as a normal cluster, if PN(i) is not located in the normal cluster, PN(i) is marked as a fluctuation abnormal suspected point, otherwise, PN(i) is not marked; If PN(i) is a fluctuation abnormal suspected point and PN(i) is a trend abnormal suspected point, PN(i) is marked as an abnormal data, otherwise, PN(i) is marked as a normal data; the marking is repeated for all discharge efficiencies in the discharge efficiency sequence; after completion, a marked efficiency sequence of the first date is obtained, which is recorded as the standard energy storage discharge data of the first date.

6. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 5, characterized in that, The abnormal analysis and processing are performed on the energy storage system discharge data and the environmental temperature data respectively to obtain standard energy storage discharge data and standard temperature data, and the following sub-steps are further included. For the environmental temperature sequence of the first date, any environmental temperature in the environmental temperature sequence is recorded as HT(j), wherein j represents a position sequence number; if HT(j) is greater than two adjacent environmental temperatures or smaller than two adjacent environmental temperatures at the same time, HT(j) is marked as a split point; all split points are repeatedly obtained, and the environmental temperature sequence is divided into multiple subsequences according to all split points; the subsequence in which HT(j) is located is recorded as a first subsequence; The first subsequence is subjected to nonlinear fitting to obtain a corresponding fitting function, which is recorded as a first fitting function, and a fitting value corresponding to HT(j) is obtained according to the first fitting function and the position corresponding to HT(j); If HT(j) is greater than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures less than the corresponding fitting value, mark HT(j) as abnormal data; if HT(j) is less than the corresponding fitting value, and there are not less than ⌈k6*k5⌉ neighborhood reference temperatures greater than the corresponding fitting value, mark HT(j) as abnormal data; otherwise, mark HT(j) as normal data, wherein k6 is a set proportion coefficient; Repeat the marking of all ambient temperatures in the ambient temperature sequence, and after completion, the corresponding marked temperature sequence is obtained, denoted as the standard temperature data of the first date. According to the standard energy storage discharge data, the efficiency calculation analysis is carried out, and the discharge efficiency prediction model is constructed according to the standard temperature data, including the following sub-steps:

7. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 6, characterized in that, Set the first time length as t2, and divide the marked efficiency sequence and the marked temperature sequence of the first date into multiple corresponding segments according to t2, respectively denoted as the divided efficiency sequence and the divided temperature sequence; Mark any segment with a length of t2 in the divided efficiency sequence as a first efficiency segment, mark any discharge efficiency in the first efficiency segment as a first efficiency PG, and set the first window size as k7; Take PG as the starting position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the first reference deviation GB1 of PG; and take PG as the middle position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the second reference deviation GB2 of PG; and take PG as the end position of the first window, and calculate the standard deviation of the charging efficiency in the first window at this time, denoted as the third reference deviation GB3 of PG; Calculate the weight deviation QG of PG, wherein QG=1 / [(GB1+GB2+GB3) / 3]; repeat the weight deviation of all discharge efficiencies in the first efficiency segment, and sum them up, denoted as the weight reference QW0; Calculate the weighted efficiency QP corresponding to PG, wherein QP=(QG / QW0)*PG, repeat the weighted efficiency corresponding to all discharge efficiencies in the first efficiency segment, and sum them up to obtain the reference discharge efficiency of the first efficiency segment; Repeat the acquisition of all reference discharge efficiencies in the divided efficiency sequence, and arrange them in corresponding order, denoted as the reference discharge efficiency sequence of the first date. According to the standard energy storage discharge data, the efficiency calculation analysis is carried out, and the discharge efficiency prediction model is constructed according to the standard temperature data, including the following sub-steps:

8. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 7, characterized in that, For the divided temperature sequence, mark any segment with a length of t2 in the divided temperature sequence as a first temperature segment, perform nonlinear fitting on the first temperature segment to obtain the corresponding fitting function, and then calculate the average temperature of the first temperature segment according to the corresponding fitting function, repeat the acquisition of all average temperatures in the divided temperature sequence, and arrange them in corresponding order, denoted as the average temperature sequence of the first date; ​ The reference discharge efficiency sequence and the average temperature sequence of the first date are recorded as the temperature efficiency data of the first date; the daily temperature efficiency data is repeatedly obtained and recorded as the temperature efficiency reference data.

9. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 8, characterized in that, The efficiency calculation analysis is performed according to the standard energy storage discharge data, and the discharge efficiency prediction model is constructed according to the standard temperature data, and the following sub-steps are further included: An original prediction model is constructed based on an LSTM model, the original prediction model includes an input layer, a hidden layer and an output layer, the output of the original prediction model is set as the reference discharge efficiency sequence of the future date; the temperature efficiency reference data is used for model training of the original prediction model, and after completion, a discharge efficiency prediction model is obtained.

10. The artificial intelligence based industrial and commercial energy storage peak shaving method according to claim 9, characterized in that, The discharge efficiency is predicted based on the discharge efficiency prediction model, and the target storage power of the energy storage system is adjusted, and the following sub-steps are included: An average temperature sequence corresponding to the future date is obtained, the discharge efficiency of the future date is predicted by using the discharge efficiency prediction model, and a reference discharge efficiency sequence of the future date is obtained; The average discharge efficiency of the future date is calculated according to the reference discharge efficiency sequence of the future date; The estimated storage power of the energy storage system for the future date is obtained, which is recorded as the target storage power of the energy storage system; The target storage power of the energy storage system is corrected by using the average discharge efficiency of the future date, the corrected storage power of the energy storage system for the future date is obtained, and the energy storage is performed according to the corrected storage power.

Images