System scheduling method and device, electronic equipment and storage medium

Abstract

The invention discloses a system scheduling method and device, electronic equipment and a storage medium, and relates to the technical field of task scheduling. The optical storage hydrogen charging system scheduling method comprises the following steps: determining current weather characteristic information and current system cooling and heating power load power; on the basis of a preset light hydrogen storage charging load power prediction model, according to the current weather characteristic information and the current system cooling and heating load power, determining cooling and heating load power at a prediction moment; determining the active power output of the optical hydrogen storage and charging system equipment based on the cold and hot electric load power at the prediction moment and the performance parameters of the optical hydrogen storage and charging system equipment; and determining an active power scheduling instruction of the optical hydrogen storage charging system equipment based on the active power output, so that the optical hydrogen storage charging system equipment controls the system state of the optical hydrogen storage charging system equipment based on the active power scheduling instruction. According to the embodiment of the invention, the active output is determined by predicting the cooling and heating power load power at the prediction moment, the active scheduling instruction is determined according to the active output, and the rationality of the active scheduling instruction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of task scheduling, and in particular to a system scheduling method, device, electronic equipment and storage medium. Background Art

[0002] The photovoltaic hydrogen storage and charging system is a comprehensive energy system that integrates photovoltaic power generation, energy storage, hydrogen production and charging. With the rapid development of new energy, the energy scheduling of the photovoltaic hydrogen storage and charging system has become a current concern.

[0003] In the existing technology, most of the energy scheduling of the photovoltaic hydrogen storage and charging system equipment is done by manual confirmation. The control results are not necessarily suitable for production, and the task scheduling is inefficient. In particular, complex scenarios combining multiple new energy sources at the enterprise level have more extensive applications, and conventional energy scheduling and control systems can no longer adapt well. Therefore, how to reasonably schedule the energy of photovoltaic hydrogen storage and charging system equipment has become a problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a system scheduling method, device, electronic device and storage medium to solve the problem of unreasonable and inefficient energy scheduling of light hydrogen storage and charging system equipment by manual determination methods in the prior art.

[0005] According to one aspect of the present invention, a system scheduling method is provided, wherein the method comprises:

[0006] Determine current weather characteristics and current system cooling, heating and electrical load power;

[0007] Based on a preset light hydrogen storage charging load power prediction model, the cold, heat and electricity load power at the prediction time is determined according to the current weather characteristic information and the current system cold, heat and electricity load power; wherein the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on a self-learning teaching and learning algorithm;

[0008] Determine the active output of the solar hydrogen storage and charging system equipment based on the cold and hot electric load power at the predicted time and the performance parameters of the solar hydrogen storage and charging system equipment;

[0009] The active power dispatching instruction of the solar hydrogen storage charging system device is determined based on the active power output, so that the solar hydrogen storage charging system device controls the system state of the solar hydrogen storage charging system device based on the active power dispatching instruction. According to another aspect of the present invention, a system dispatching device is provided, wherein the device comprises:

[0010] A data determination module is used to determine the current weather characteristic information and the current system cooling, heating and electric load power;

[0011] A power prediction module is used to determine the cold, heat and electric load power at the prediction time based on a preset light hydrogen storage charging load power prediction model, according to the current weather characteristic information and the cold, heat and electric load power of the current system; wherein the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on a self-learning teaching and learning algorithm;

[0012] An output determination module, used to determine the active output of the solar hydrogen storage charging system equipment based on the cold and hot electric load power at the predicted time and the performance parameters of the solar hydrogen storage charging system equipment;

[0013] An instruction generation module is used to determine the active scheduling instruction of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction.

[0014] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the system scheduling method described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the system scheduling method described in any embodiment of the present invention when executed.

[0019] The technical solution of the embodiment of the present invention determines the current weather characteristic information and the current system cold, hot and electric load power, based on a preset solar hydrogen storage charging load power prediction model, and determines the cold, hot and electric load power at the prediction time according to the current weather characteristic information and the current system cold, hot and electric load power, so as to achieve accurate prediction of the cold, hot and electric load power at the prediction time; determines the active output of the solar hydrogen storage charging system equipment based on the cold, hot and electric load power at the prediction time and the performance parameters of the solar hydrogen storage charging system equipment, and determines the active scheduling instructions of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instructions, thereby achieving determination of the active scheduling instructions by the active output, improving the rationality of the active scheduling instructions, and ensuring the reasonable distribution of power inside the solar hydrogen storage charging system equipment.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flowchart of a system scheduling method provided according to Embodiment 1 of the present invention;

[0023] Figure 2 is a flowchart of a system scheduling method provided according to Embodiment 2 of the present invention;

[0024] Figure 3 is a flowchart of a system scheduling method provided according to Embodiment 3 of the present invention;

[0025] Figure 4 is a training flow chart of a preset light hydrogen storage charging load power prediction model provided in Embodiment 3 of the present invention;

[0026] Figure 5 is a schematic diagram of the architecture of a preset light hydrogen storage charging load power prediction model provided according to the third embodiment of the present invention;

[0027] Figure 6 is a structural diagram of a system scheduling device provided according to Embodiment 4 of the present invention;

[0028] Figure 7 It is a structural schematic diagram of an electronic device for implementing a system scheduling method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] Embodiment 1

[0032] Figure 1 This is a flow chart of a system scheduling method provided according to the first embodiment of the present invention. This embodiment can be applied to determine the active scheduling instructions of the photovoltaic hydrogen storage charging system equipment and schedule the photovoltaic hydrogen storage charging system equipment. The method can be executed by a system scheduling device. The system scheduling device can be implemented in the form of hardware and / or software. The system scheduling device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0033] S110, determining current weather characteristic information and current system cooling, heating and electric load power.

[0034] Among them, the current weather characteristic information can be understood as an indicator that characterizes the current weather conditions. Exemplarily, the current weather characteristic information may include the daily average light intensity, the daily average temperature, the daily average humidity, the daily average cloud cover, the daily maximum temperature and the daily minimum temperature. In the actual operation process, the weather characteristic information can be in the form of a characteristic vector, which is convenient for the preset light storage hydrogen charging load power prediction model to analyze and process the weather characteristic information. The current system cold, hot and electric load power can be understood as the cold load power, heat load power and electric load power of the light storage hydrogen charging system equipment at the current moment. Among them, the cold load power is the refrigeration power required to maintain the temperature in a specific space lower than the ambient temperature, the heat load power is the heating power required to maintain the temperature in a specific space higher than the ambient temperature, and the electric load power is the total electric power consumed by the light storage hydrogen charging system equipment.

[0035] In an embodiment, weather information may be determined first, and current weather characteristic information may be determined according to the weather information. For example, light intensity, temperature, humidity, cloud cover, temperature, and temperature at multiple times may be determined, and then the daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature, and daily minimum temperature may be determined according to the weather information as weather characteristic information. Then the system cooling, heating, and electric load power at the current time is detected, and the current time is used as the system cooling, heating, and electric load power.

[0036] S120, based on the preset solar hydrogen storage charging load power prediction model, determine the cold, heat and electricity load power at the prediction time according to the current weather characteristic information and the current system cold, heat and electricity load power.

[0037] Among them, the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on the self-learning teaching and learning algorithm. The preset light hydrogen storage charging load power prediction model can be understood as a pre-trained model for predicting the cold, heat and electricity load power at the prediction moment. The cold, heat and electricity load power at the prediction moment can be understood as the cold, heat and electricity load power at the preset prediction moment, and the prediction moment can be multiple consecutive moments after the current moment.

[0038] In an embodiment of the invention, the current weather characteristic information and the current system cold, heat and electricity load power can be input into a preset light hydrogen storage charging load power prediction model, and the cold, heat and electricity load power at the prediction time can be determined by the preset light hydrogen storage charging load power prediction model. In the actual operation process, the preset light hydrogen storage charging load power prediction model can determine the weather category based on the current weather characteristic information, and match the cold, heat and electricity load power of the historical day corresponding to the current system cold, heat and electricity load power, and determine the cold, heat and electricity load power at the prediction time based on the cold, heat and electricity load power of the historical day as the cold, heat and electricity load power at the prediction time corresponding to the current weather characteristic information.

[0039] S130. Determine the active output of the solar hydrogen storage and charging system equipment based on the cold and hot electric load powers and the performance parameters of the solar hydrogen storage and charging system equipment at the predicted time.

[0040] Among them, the photovoltaic hydrogen storage and charging system equipment can be understood as a comprehensive energy system equipment that integrates photovoltaic power generation, energy storage, hydrogen production and charging functions. For photovoltaic hydrogen storage and charging system equipment, there are cold, hot and electric load powers. The performance parameters can be understood as parameters that characterize the performance of the photovoltaic hydrogen storage and charging system equipment. Exemplarily, the performance parameters at least include the refrigeration coefficient of the photovoltaic hydrogen storage and charging system equipment and the thermal efficiency of the photovoltaic hydrogen storage and charging system equipment. The active output can be understood as the active power actually output by the photovoltaic hydrogen storage and charging system equipment.

[0041] In an embodiment, the cold load power at the predicted moment, the heat load power at the predicted moment, and the electric load power at the predicted moment included in the cold, hot and electric load power at the predicted moment can be determined. Then the ratio of the heat load power at the predicted moment to the thermal efficiency is determined as the active output corresponding to the heat load power at the predicted moment, the ratio of the cold load power at the predicted moment to the cooling coefficient is determined as the active output corresponding to the cold load power at the predicted moment, the electric load power at the predicted moment is determined as the active output corresponding to the electric load power at the predicted moment, the active output corresponding to the heat load power at the predicted moment, the active output corresponding to the cold load power at the predicted moment, and the sum of the active output corresponding to the electric load power at the predicted moment, the sum is used as the active output of the solar hydrogen storage charging system equipment.

[0042] S140. Determine an active scheduling instruction of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction.

[0043] Among them, the active power dispatching instruction can be understood as the instruction information for dispatching the power of the photovoltaic hydrogen storage and charging system equipment. Among them, the active power dispatching instruction can include negative active power dispatching instructions and positive active power dispatching instructions, and the active power dispatching instruction can be determined according to the active power output. The system state of the photovoltaic hydrogen storage and charging system equipment can be understood as the state of the photovoltaic hydrogen storage and charging system equipment. Exemplarily, the system state of the photovoltaic hydrogen storage and charging system equipment can at least include the charge state of the energy storage battery and the hydrogen storage capacity of the hydrogen fuel cell.

[0044] In an embodiment, an active output threshold can be pre-configured. When it is determined that the active output is greater than the preset active output threshold, a negative active dispatch instruction is generated; when it is determined that the active output is less than the preset active output threshold, a positive active dispatch instruction is generated to adjust the system state of the photovoltaic hydrogen storage charging system equipment until the active output is equal to the active output threshold. In the actual operation process, the system state of the photovoltaic hydrogen storage charging system equipment can be controlled by multiple aspects, such as adjusting the energy storage battery charging load power, the hydrogen fuel cell cold charge power, the hydrogen fuel cell hot charge power, the energy storage battery discharge load power, the purchase of electricity from the power grid, the sale of electricity to the power grid, the controllable cold load power, the controllable heat load power and the controllable electric load power, etc. The adjustment of each electric load power, heat load power and cold load power can be based on the active dispatch instruction. The active dispatch instruction corresponding to each electric load power, heat load power and cold load power can be generated according to the active output adaptability, as long as the control amount increment of the photovoltaic hydrogen storage charging system equipment is minimized.

[0045] The embodiment of the present invention determines the current weather characteristic information and the cold, hot and electric load power of the current system based on a preset solar hydrogen storage charging load power prediction model, determines the cold, hot and electric load power at the prediction time according to the current weather characteristic information and the cold, hot and electric load power of the current system, and realizes accurate prediction of the cold, hot and electric load power at the prediction time; determines the active output of the solar hydrogen storage charging system equipment based on the cold, hot and electric load power at the prediction time and the performance parameters of the solar hydrogen storage charging system equipment, and determines the active scheduling instructions of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instructions, realizes determining the active scheduling instructions through the active output, improves the rationality of the active scheduling instructions, and ensures the reasonable distribution of power inside the solar hydrogen storage charging system equipment.

[0046] In one embodiment, the initial training process of the preset photovoltaic hydrogen storage charging load power prediction model includes:

[0047] Obtain weather characteristic information of historical days, weather characteristic information of the day to be tested, cold, heat and electricity load power of historical days, and cold, heat and electricity load power of the day to be tested, and perform data preprocessing on the weather characteristic information of historical days and the cold, heat and electricity load power of historical days associated with the weather characteristic information of each historical day;

[0048] Cluster the weather characteristic information of historical days according to the K-means clustering algorithm, and mark the weather category of the weather characteristic information of each historical day;

[0049] According to the weather characteristic information of the day to be measured, the weather category to which the weather characteristic information of the day to be measured belongs is determined as the first category, and the cooling, heating and electric load power of the historical day associated with the weather characteristic information of the historical day corresponding to the first category is extracted as the cooling, heating and electric load power to be matched;

[0050] Determine the power similarity between each to-be-matched cold, heat and electricity load power and the cold, heat and electricity load power of the day to be measured, arrange the power similarities in descending order, and take the to-be-matched cold, heat and electricity load power corresponding to the preset number of power similarities arranged in front as the target cold, heat and electricity load power of the day to be measured, and take the historical day to which the target cold, heat and electricity load power belongs as the similar day;

[0051] Determine the cooling, heating and electricity load powers at the historical moments of the preset sliding window width before the similar day as the historical cooling, heating and electricity load powers;

[0052] The training set and the test set are formed according to the weather characteristic information of the test day, the cooling, heating and electricity load power of the test day, the historical cooling, heating and electricity load power, and the target cooling, heating and electricity load power;

[0053] The training set is input into the preset solar hydrogen storage charging load power prediction model, and the optimal weight and the optimal threshold are determined based on the self-learning teaching and learning algorithm until the preset number of iterations is reached, and the preset solar hydrogen storage charging load power prediction model is trained based on the optimal weight and the optimal threshold.

[0054] Among them, the weather characteristic information of the historical day can be the historical weather characteristic information; the weather characteristic information of the day to be tested can be understood as the weather characteristic information of the day to be trained; the cold, heat and electricity load power of the day to be tested can be understood as the cold, heat and electricity load power of the day to be trained. In the actual operation process, the weather characteristic information of the historical day and the weather characteristic information of the day to be tested can both be the weather characteristic information of the previous year, as long as the day to be tested is located at the rear moment of the historical day. K-means clustering algorithm is an unsupervised learning algorithm, which is mainly used to divide data points into K different clusters, and can cluster the weather characteristic information of the historical day into different weather categories, wherein the weather category can at least include sunny, cloudy, overcast and expected rain and snow. The preset sliding window width number can be understood as a preset number of the same size as the sliding window width, so that the cold, heat and electricity load power at the historical moment of the preset sliding window width number is used as an independent input sample for the training of the preset light storage hydrogen charging load power prediction model. In one embodiment, the preset sliding window width number can be pre-set according to user needs. For example, the preset sliding window width number can include but is not limited to 4, 5, 6, etc. The sliding window width refers to the window size used to divide data when processing time series data or continuous data. The preset number of iterations may be a preset number of training iterations, which may be set according to user needs. Exemplarily, the preset number of iterations may include but is not limited to 150, 200, 250, etc.

[0055] In an embodiment, the weather characteristic information of the day to be measured, the cold, heat and electricity load power of the day to be measured, the weather characteristic information of the historical day, and the cold, heat and electricity load power of the historical day associated with the weather characteristic information of each historical day can be obtained to perform data preprocessing on the weather characteristic information of the historical day and the cold, heat and electricity load power of the historical day associated with the weather characteristic information of each historical day. In the actual operation process, the average values ​​of the weather characteristic information of the historical day and the cold, heat and electricity load power of the historical day can be determined respectively, and the weather characteristic information of the historical day and the cold, heat and electricity load power of the historical day whose deviation from the average value is greater than three times the standard deviation can be determined as abnormal values ​​respectively, and the abnormal values ​​are eliminated, and then the cubic spline difference is performed to ensure the integrity of the weather characteristic information of the historical day and the cold, heat and electricity load power of each historical day. Then, the processed weather characteristic information of the historical day and the cold, heat and electricity load power of each historical day are normalized to ensure that the weather characteristic information of the historical day and the cold, heat and electricity load power of each historical day are in the same numerical range.

[0056] Then cluster the weather characteristic information of historical days through the K-means clustering algorithm, and mark the weather category of the weather characteristic information of each historical day. Determine the weather category corresponding to the weather characteristics of the day to be tested, obtain the cold, heat and electricity load power of the historical day associated with the weather characteristic information of the historical day corresponding to the first category as the cold, heat and electricity load power to be matched, determine the power similarity between each cold, heat and electricity load power to be matched and the cold, heat and electricity load power of the day to be tested, arrange the power similarity in order from large to small, and take the cold, heat and electricity load power to be matched corresponding to the preset number of power similarities arranged in front as the target cold, heat and electricity load power corresponding to the cold, heat and electricity load power of the day to be tested, and take the historical day to which the target cold, heat and electricity load power belongs as a similar day. In the actual operation process, the cosine similarity method can be used to determine the power similarity between the cold, heat and electricity load power to be matched and the cold, heat and electricity load power of the day to be tested. The cold, hot and electric load power of the preset sliding window width number of historical moments before the similar day is extracted as the historical cold, hot and electric load power, the weather characteristic information of the day to be tested and the cold, hot and electric load power of the day to be tested are used as the input of the model, and the target cold, hot and electric load and the historical cold, hot and electric load power are used as the output of the model to form a training set and a test set. Among them, the ratio of the training set to the test set can be preset. For example, the ratio of the training set to the test set can be 7:3, 8:2, 9:1, etc. The training set is then input into the preset light storage hydrogen charging load power prediction model, and the optimal weight and the optimal threshold are determined by the self-learning teaching and learning algorithm until the preset number of iterations is reached, and the preset light storage hydrogen charging load power prediction model is trained by the optimal weight and the optimal threshold. In one embodiment, after the preset light storage hydrogen charging load power prediction model is trained based on the optimal weight and the optimal threshold, the output of the model can also be placed at the end of the input data queue as the input of the new preset light storage hydrogen charging load power prediction model, so as to realize the reverse propagation of the preset light storage hydrogen charging load power prediction model and improve the accuracy of the preset light storage hydrogen charging load power prediction model. In one embodiment, the number of hidden layer neurons of the preset light hydrogen storage charging load power prediction model can be 4, the preset sliding window width can be set to 5, the population size of the self-learning teaching and learning algorithm can be 50, and the self-learning weight can be 3.

[0057] In one embodiment, the training set is input into a preset solar hydrogen storage charging load power prediction model, and the optimal weight and threshold are determined based on a self-learning teaching and learning algorithm until a preset number of iterations is reached, including:

[0058] Initialize a preset number of student populations; each population represents a set of weights and thresholds;

[0059] Based on each student population, the training set is input into the preset light storage hydrogen charging load power prediction model to determine the mean square error corresponding to each student population;

[0060] The student population corresponding to the minimum mean square error among the mean square errors is taken as the teacher population, and the corresponding weights and thresholds of each student population are updated according to the teacher population and the preset self-learning weights until the preset number of iterations is reached, and the weights and thresholds corresponding to the teacher population are determined as the optimal weights and optimal thresholds.

[0061] The student population refers to a set of randomly initialized weights and thresholds. In actual operation, the number of student populations can be a preset number. The preset number refers to the number of student populations set in advance. For example, the preset number can include but is not limited to 45, 50 or 55. The mean square error is used to measure the average square difference between the predicted value and the actual value of the weight and threshold corresponding to the training represented by the student population. For example, the mean square error can be Among them, y real，i is the expected value, i.e. the target cooling, heating and electricity load; pre，i is the predicted value, i.e., the input value of the preset light hydrogen storage charging load power prediction model; N is the total number of samples, i.e., the number of input data of the preset light hydrogen storage charging load power prediction model; i represents the i-th sample. The preset self-learning weight refers to the preset step size and direction of the control student population in the process of learning from the teacher population. For example, the preset self-learning weight can be 3, 4, 5, etc.

[0062] In an embodiment, a preset number of weights and thresholds can be initialized, and each set of weights and thresholds can be used as a student population. For each student population, the training set is input into the preset light storage hydrogen charging load power prediction model to obtain the output value (prediction value) of the model, determine the mean square error corresponding to each student population, and determine the mean square error corresponding to each student population according to the target cold, hot and electric load (expected value) corresponding to each input value. The student population corresponding to the mean square error with the smallest value in each mean square error is used as the teacher population. Each student population updates the corresponding weights and thresholds according to the teacher population and the preset self-learning weights, iterates to the preset number of iterations, ends the optimization process, and uses the weights and thresholds corresponding to the final teacher population as the optimal weights and optimal thresholds. In the actual process, the updating method of the weights and thresholds can include adding an adjustment amount related to the self-learning weights and the difference between the teacher population and the student population on the basis of the weights and thresholds of the current student population. So that the student population gradually approaches the teacher population and improves its prediction accuracy.

[0063] In one embodiment, after the preset light hydrogen storage charging load power prediction model is trained based on the optimal weight and the optimal threshold, the method further includes:

[0064] The weather characteristic information of the day to be tested and the cold, heat and electricity load power of the day to be tested are input into the trained preset light storage hydrogen charging load power prediction model to obtain the predicted cold, heat and electricity load power;

[0065] Determine the deviation value between the predicted cooling, heating and electric load power and the target cooling, heating and electric load power;

[0066] The parameters of the preset photo-hydrogen storage charging load power prediction model are adjusted according to the deviation value until the deviation value reaches the target value.

[0067] In an embodiment, since the prediction result of the cold, heat and electricity load power at the prediction moment cannot guarantee that its actual value is the same as the predicted value, which leads to a deviation between the power output issued in advance and the actual active output, correction is required. The weather characteristic information of the day to be measured and the cold, heat and electricity load power of the day to be measured can be input into the preset light storage hydrogen charging load power prediction model that has been trained to obtain the predicted cold, heat and electricity load power, that is, the predicted value, and determine the deviation value between the predicted cold, heat and electricity load power and the target cold, heat and electricity load power. At this time, the parameters of the preset light storage hydrogen charging load power prediction model can be adaptively adjusted according to the deviation value until the deviation value reaches the target value. Exemplarily, the threshold and weight of the preset light storage hydrogen charging load power prediction model can be adjusted. Among them, the target value can be set by the user based on the actual situation. Exemplarily, the target value can include but is not limited to 0. At the same time, the cold, heat and electricity load power at the next moment can be used as the initial value of the next round of rolling optimization of the preset light storage hydrogen charging load power prediction model to optimize the preset light storage hydrogen charging load power prediction model. In one embodiment, the actual cold, heat and electricity load power of the photovoltaic hydrogen storage system equipment at the current moment can be used as the initial value for a new round of optimization to reduce the deviation between the predicted value and the actual value, thus overcoming the uncertainty of the system and photovoltaics. By comparing the predicted processing values ​​of each device at the current moment, the interference value can be obtained, namely: init =u(k)+δΔu(k), where u(k) represents the output value determined by the predicted cooling, heating and electric load power of each system equipment at time k; u init It represents the actual output value of each system equipment at time k and serves as the initial value at time k+1. δ represents the size of the interference, which can be 5%, 10%, or 15%. The cold, hot, and electric load power at the next moment can be used for further rolling optimization until the output value determined by the cold, hot, and electric load power predicted by the system equipment is the same as the actual output value of the system equipment.

[0068] Embodiment 2

[0069] Figure 2 1 is a flow chart of a system scheduling method provided according to Embodiment 2 of the present invention. This embodiment is a further description of a system scheduling method based on the above embodiment. Figure 2 As shown, the method includes:

[0070] S210. Determine weather information, and determine current weather characteristic information according to the weather information; wherein the current weather characteristic information at least includes daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature, and daily minimum temperature.

[0071] The weather information may be understood as a description of the weather conditions within a specific time. For example, the weather information may include but is not limited to light intensity, temperature, humidity, and cloud cover.

[0072] In an embodiment, weather information for multiple time periods of a day can be obtained, and the daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature and daily minimum temperature can be determined according to the weather information, and the daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature and daily minimum temperature can be used as current weather characteristic information.

[0073] S220, detecting the system cooling, heating and electricity load power at the current moment as the current system cooling, heating and electricity load power.

[0074] In an embodiment, the current system heat and cold electric load power can be determined by an electric energy meter, and the current system heat and cold electric load power is used as the current system heat and cold electric load power. Alternatively, the current and voltage of the solar hydrogen storage charging system equipment can be determined by an ammeter and a voltmeter to calculate the current system heat and cold electric load power.

[0075] S230. Based on the preset solar hydrogen storage charging load power prediction model, determine the cold, heat and electricity load power at the prediction time according to the current weather characteristic information and the current system cold, heat and electricity load power.

[0076] S240, determining the performance parameters of the solar hydrogen storage and charging system equipment, and extracting a preset active output confirmation algorithm.

[0077] Among them, the performance parameters at least include the refrigeration coefficient of the photovoltaic hydrogen storage charging system equipment and the thermal efficiency of the photovoltaic hydrogen storage charging system equipment. The preset active output confirmation algorithm can be understood as a preset algorithm for determining the active output of the photovoltaic hydrogen storage charging system equipment. Exemplarily, the preset active output confirmation algorithm can be to calculate the active output corresponding to the heat load power, the active output corresponding to the cold load power, and the electric load power of the photovoltaic hydrogen storage charging system equipment respectively, and take the sum of the active output corresponding to the heat load power, the active output corresponding to the cold load power, and the electric load power as the active output of the photovoltaic hydrogen storage charging system equipment.

[0078] In an embodiment, the refrigeration coefficient of the photovoltaic hydrogen storage and charging system equipment and the thermal efficiency of the photovoltaic hydrogen storage and charging system equipment can be extracted, and a pre-set preset active output confirmation algorithm can be extracted.

[0079] S250, determining the active output of the photovoltaic hydrogen storage and charging system equipment according to the preset active output confirmation algorithm, performance parameters and the cold and hot electric load powers at the predicted time.

[0080] In an embodiment, the active output corresponding to the cold load power can be determined by the cold electric load power at the preset time and the cooling coefficient in the cold and hot electric load power at the predicted time according to the preset active output confirmation algorithm; the active output corresponding to the heat load power can be determined by the heat electric load power at the preset time and the thermal efficiency in the cold and hot electric load power at the predicted time; the electric load power at the predicted time in the cold and hot electric load power at the predicted time can be used as the active output corresponding to the electric load power. In the actual operation process, the ratio of the cold electric load power at the preset time to the cooling coefficient can be determined as the active output corresponding to the cold load power, the ratio of the heat electric load power at the preset time to the thermal efficiency can be used as the active output corresponding to the heat load power, and the sum of the active output corresponding to the cold load power, the active output corresponding to the heat load power, and the active output corresponding to the electric load power can be determined as the active output of the solar hydrogen storage and charging system equipment.

[0081] S260: Extract a pre-configured active output threshold.

[0082] Among them, the active output threshold can be understood as the preset active output critical value of the photovoltaic hydrogen storage charging system equipment. Since the active output is the sum of the positive output and the negative output, the active output threshold required by the photovoltaic hydrogen storage charging system equipment can be determined according to user needs. For example, the active output threshold can also include 0, that is, the positive output and the negative output are the same.

[0083] S270. When it is determined that the active output is greater than the preset active output threshold, a negative active dispatch instruction is generated to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold.

[0084] In an embodiment, when it is determined that the active output is greater than a preset active output threshold, it can be considered that the positive output of the photovoltaic hydrogen storage charging system device is high, and a negative active scheduling instruction can be generated to adjust the system state of the photovoltaic hydrogen storage charging system device until the active output is equal to the active output threshold. In one embodiment, the system state of the photovoltaic hydrogen storage charging system device can be adjusted by increasing the power consumption of the controllable load or reducing hydrogen power generation through a negative active scheduling instruction.

[0085] S280. When it is determined that the active output is less than the preset active output threshold, a forward active dispatch instruction is generated to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold.

[0086] In an embodiment, when it is determined that the active output is less than a preset active output threshold, it can be considered that the negative output of the photovoltaic hydrogen storage charging system device is high, and a positive active dispatch instruction can be generated to adjust the system state of the photovoltaic hydrogen storage charging system device until the active output is equal to the active output threshold. In an embodiment, the system state of the photovoltaic hydrogen storage charging system device can be adjusted by reducing the power consumption of the controllable load or increasing hydrogen power generation through a positive active dispatch instruction.

[0087] In an embodiment of the present invention, weather information is determined, and current weather characteristic information is determined according to the weather information; wherein the current weather characteristic information includes at least daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature and daily minimum temperature; the system cold, heat and electricity load power at the current moment is detected as the current system cold, heat and electricity load power; based on a preset light hydrogen storage charging load power prediction model, the cold, heat and electricity load power at the prediction moment is determined according to the current weather characteristic information and the current system cold, heat and electricity load power; the cold, heat and electricity load power at the future moment is predicted according to the current weather characteristic information and the current system cold, heat and electricity load power; by determining the performance parameters of the light hydrogen storage charging system equipment, and extracting the preset active power output power confirmation algorithm; determine the active output of the photovoltaic hydrogen storage charging system equipment according to the preset active output confirmation algorithm, performance parameters and the cold and hot power load power at the predicted time to ensure the accuracy of the active output calculation; by extracting the pre-configured active output threshold, when it is determined that the active output is greater than the preset active output threshold, generate a negative active dispatch instruction to adjust the system state of the photovoltaic hydrogen storage charging system equipment until the active output is equal to the active output threshold; when it is determined that the active output is less than the preset active output threshold, generate a positive active dispatch instruction to adjust the system state of the photovoltaic hydrogen storage charging system equipment until the active output is equal to the active output threshold, thereby realizing the dispatch of the active output of the photovoltaic hydrogen storage charging system equipment and achieving energy dispatch balance.

[0088] Embodiment 3

[0089] Figure 3 is a flow chart of a system scheduling method provided according to Embodiment 3 of the present invention. Figure 3 As shown, this embodiment is based on the above embodiment, taking the historical day as the previous year as an example, to further illustrate a system scheduling method. Figure 3 As shown, the method includes:

[0090] S310, establish a preset solar hydrogen storage charging load power prediction model based on a back propagation neural network optimized by a self-learning teaching and learning algorithm, input the weather characteristic information of the previous year and the cold and heat power load power of the previous year, and complete the training of the preset solar hydrogen storage charging load power prediction model.

[0091] In an embodiment, the weather characteristic information of the previous year and the cooling, heating and electric load power of the previous year may be subjected to data preprocessing. For example, the average value and standard deviation of the weather characteristic information of the previous year and the average value and standard deviation of the cooling, heating and electric load power of the previous year may be determined respectively, and the missing values ​​and abnormal values ​​of the weather characteristic information of the previous year and the cooling, heating and electric load power of the previous year whose deviation from the average value exceeds three times the standard deviation 3σ are removed, and then cubic spline difference interpolation is performed to ensure the integrity of the array. Among them, σ is the standard deviation, N is the total number of samples (the weather characteristic information of the previous year or the number of cold, heat and electricity load power of the previous year), μ is the individual mean in the sample (the average value of the weather characteristic information of the previous year or the cold, heat and electricity load power of the previous year), x i The sample points are (each weather characteristic information of the previous year or the cooling, heating and electricity load power of the previous year). Then the processed weather characteristic information of the previous year and the cooling, heating and electricity load power of the previous year are respectively subjected to data normalization. Among them, y max ,y min is the maximum and minimum value of the normalized interval, x max , x min are the maximum and minimum values ​​of the sample, x is the current data (the weather characteristic information of the previous year or the cooling, heating and electricity load power of the previous year), and y is the normalized data (the normalized weather characteristic information of the previous year or the cooling, heating and electricity load power of the previous year).

[0092] For different weather conditions, the power curve of photovoltaic power generation is different. We can use the K-means clustering algorithm to extract the weather information feature vector (weather feature information) of the previous year. res Clustering is performed to divide the historical data (the weather characteristic information of the previous year) into four categories: sunny, cloudy, overcast, and expected rain and snow. Among them, the weather information feature vector (weather characteristic information) can be WI res =[G m , T m , Rh m , Cd m , T h , T l ]; where G m is the average daily light intensity, T m is the daily average temperature, Rh m is the daily average humidity, Cd m is the daily average cloud cover, T h is the maximum daily temperature, T l The lowest temperature of the day.

[0093] In order to obtain the similar days corresponding to the cooling, heating and electricity load data, the characteristic vectors of cooling load power, heating load power and electricity load power (the cooling, heating and electricity load power of the previous year) WI can be selected respectively. C , WI H , WI E . Based on the weather information feature vector WI res , cooling load power, heating load power, electrical load power characteristic vector WI C , WI H , WI E Calculate the similarity between the day to be tested and the historical days, and select the top N days with the highest similarity as the similar days of the day to be predicted by comparison and sorting.

[0094] Among them, the cooling load characteristic vector is WI C =[Q C (1), Q C (2), …, Q C (i), …, Q C (L)];

[0095] The heat load characteristic vector is WI H =[Q H (1), Q H (2), …, Q H (i), …, Q H (L)];

[0096] The electric load characteristic vector is WI E =[P E (1), P E (2), …, P E (i), …, P E (L)]; where Q is the cooling and heating load power, and P is the electrical load power.

[0097] In actual operation, the similarity can be calculated by cosine similarity. The larger the cosine similarity value, the higher the similarity, and vice versa. Where A and B are two vectors (respectively, the cooling, heating and electric load power of each previous year and the cooling, heating and electric load power of the day to be measured), A·B represents their dot product (inner product), and ||A|| and ||B|| represent their norms (modules).

[0098] In one embodiment, Figure 4 1 is a training flow chart of a preset light hydrogen storage charging load power prediction model provided in Embodiment 3 of the present invention. Figure 4As shown, the original data (the weather characteristic information of the day to be tested, the cold and hot electric load power of the day to be tested, the historical cold and hot electric load power and the target cold and hot electric load power constitute the training set) can be processed to generate a training set and a test set; determine the structure of the back propagation neural network (preset light storage hydrogen charging load power prediction model), that is, the number of neurons in the input, implicit and output layers. Exemplarily, the weather characteristic information of the day to be tested and the cold and hot electric load power of the day to be tested can be used as the input of the model, and the target cold and hot electric load and the historical cold and hot electric load power can be used as the output of the model, and the number of neurons in the hidden layer of the back propagation neural network is 4. Then set the parameters of the self-learning teaching and learning algorithm, such as the sliding window width q is 5, the population size of the self-learning teaching and learning optimization is 50, the preset number of iterations is 200, and the self-learning weight ω is 3. Generate an initial student population and calculate its fitness function value and population mean, and select the individual with the best fitness as the teacher. In this way, the optimization of the teaching stage, the learning stage, and the self-learning stage is carried out. The global optimal individual is determined by the change of fitness and the new population mean is calculated. When the number of iterations reaches 200, the iteration is confirmed to be completed, the optimal individual is output, and the optimal weight and the optimal threshold are obtained. Based on the optimal weight and the optimal threshold, the back propagation neural network (the preset light hydrogen storage charging load power prediction model) is trained, and the preset light hydrogen storage charging load power prediction model obtained by training is used to make a prediction to obtain the predicted value of the prediction day.

[0099] Among them, the self-learning teaching and learning algorithm is as follows:

[0100] The teaching stage is the process of teachers sharing their knowledge with students. First, the fitness function values ​​are sorted and the one with the best fitness function value is selected from the entire population as the teacher. Each student in the class will obtain knowledge from the teacher and update it as follows:

[0101]

[0102] In the formula, are the jth new and original students in the i-th iteration respectively. The new student competes with its predecessor. When its fitness function value is better than the original student, the new student will replace the original student. i is the difference between the teacher and student means in the i-th iteration; Mean i is the mean of the students; T i For the selected teacher; T F is the teaching factor, and its value range is (1, 2).

[0103] After the teaching phase, the learning phase provides each student with an opportunity to learn from their classmates. In this phase, each student will randomly select another student to compare fitness and interactively update the knowledge storage based on the size of fitness. This phase is represented by the following formula:

[0104]

[0105] Where, X ij , X ik In the ith iteration, the jth and kth students are randomly selected from the class respectively; the original students will update their knowledge according to the difference between the two, and the above two formulas can be used to update them according to the size of their fitness function values. After the update is completed, it is necessary to choose whether to replace or discard them according to the size of the fitness value.

[0106] The self-learning stage aims to further explore the space near the particle position in order to obtain the global optimal solution. Each student has the opportunity to learn from the space around him. The self-learning stage is expressed as follows:

[0107] In the formula, ω is the self-learning weight, which determines the self-learning range of each student, and its sampling range is (0.1, 10).

[0108] The fitness function is calculated by calculating the mean square error between the predicted value and the expected value obtained by each particle. The calculation formula is as follows: Among them, y real，i is the expected value, i.e. the target cooling, heating and electricity load; pre，i is the predicted value, i.e., the input value of the preset light hydrogen storage charging load power prediction model; N is the total number of samples, i.e., the number of input data of the preset light hydrogen storage charging load power prediction model; i represents the i-th sample.

[0109] After completing the optimization of the particle teaching phase, the learning phase, and the self-learning phase, it is determined whether the preset number of iterations has been reached. If not, the above process is repeated. If completed, the optimization process ends.

[0110] In one embodiment, the data sequence of the solar hydrogen storage charging / cold and hot electricity load power and the weather information data at the current time before the test day can be input. Where q is a sliding window, i = 1, 2, ..., 7, representing the photovoltaic hydrogen storage and cooling, heating and electricity loads respectively. Substitute it into the trained light hydrogen storage charging load power prediction model to get the predicted value at the current moment You can Put in The tail of the queue, and remove the value at the head of the queue as the new input Repeat the input until the prediction time domain length p is reached, where p is the prediction step length.

[0111] In one embodiment, Figure 5 Schematic diagram of the architecture of the preset light hydrogen storage charging load power prediction model provided according to the third embodiment of the present invention. Figure 5 As shown, the preset light hydrogen storage charging load power prediction model includes an input layer, a hidden layer and an output layer. The input layer can input the cold, hot and electric load power of the light hydrogen storage charging system equipment, and the output can be the cold, hot and electric load power of the hydrogen storage charging system equipment. The output layer can compare the predicted output with the actual value for error back propagation to optimize the preset light hydrogen storage charging load power prediction model.

[0112] S320, based on the preset solar hydrogen storage charging load power prediction model, predict the solar hydrogen storage charging power at the predicted moments of multiple consecutive moments through the current weather information and current power data, and determine the active output of the solar hydrogen storage charging system equipment.

[0113] In one embodiment, after the training of the preset solar hydrogen storage charging load power prediction model is completed, feedback correction optimization can be performed on the solar hydrogen storage charging power and the active output of the solar hydrogen storage charging system equipment at the prediction moment and the current actual solar hydrogen storage charging power and the active output of the system equipment to eliminate the uncertainty of the solar hydrogen storage charging power and the active output of the system equipment and obtain a predicted value with a reduced deviation from the actual value.

[0114] In the embodiment, since the prediction results of the photovoltaic hydrogen storage charging and the cold and heat power load cannot guarantee that their actual values ​​are the same as the predicted values, which leads to a deviation between the power output issued in advance and the actual active power output, a correction link is required. The actual photovoltaic hydrogen storage charging cold and heat power load power and the system active power output at the current moment are used as the initial values, and a new round of optimization is carried out to reduce the deviation between the predicted value and the actual value, thereby overcoming the uncertainty of the system and photovoltaics. That is:

[0115] u init =u(k)+δΔu(k);

[0116] x init = x(k+1);

[0117] Where: u(k) represents the predicted output value of each device at time k; u init represents the actual output value of each device at time k and serves as the initial value at time k+1. δ represents the magnitude of the interference, which can be 5%, 10%, or 15%. x(k+1) represents the actual value of the system state at time k+1 after the actual output value of each device at time k is sent. The actual value of the system state at time k+1 can be used as the initial value for the next rolling optimization, that is, x init =x(k+1); where x init Indicates the initial value of the system state quantity at time k+1. Based on the above prediction values, solar hydrogen storage charging, load power and the latest weather information, multiple continuous active output instructions of the solar hydrogen storage charging system equipment can be determined.

[0118] S330. Based on the cold and hot electric load power at the predicted time and the latest weather information, determine a plurality of continuous active power output instructions of the photovoltaic hydrogen storage charging system equipment, so that the photovoltaic hydrogen storage charging system equipment controls the system state of the photovoltaic hydrogen storage charging system equipment based on the active power dispatching instructions.

[0119] In one embodiment, the photovoltaic hydrogen storage and charging system equipment can be modeled, with the state of charge of the energy storage battery and the hydrogen storage capacity of the hydrogen fuel cell as state variables and output variables. The hydrogen fuel cell, energy storage battery, grid interaction power, and controllable load output of cold and hot electricity are used as control variables to form a system model. The discrete state space equation is as follows:

[0120]

[0121] Where x = [SOC BT , HS FC ];

[0122] y=[SOC BT , HS FC ];

[0123] u=

[0124] [P BT，chr , P BT，dis , P grid , P excess , Q FC，chr , Q FC，dis , Q CL，C , Q CL，H , Q CL，E ]

[0125]

[0126]

[0127]

[0128] Among them, the variables are x, which is the state variable of electric energy storage and hydrogen energy storage; y, which is the output variable of electric energy storage and hydrogen energy storage; A, which is the energy ratio of electric energy storage and hydrogen energy storage; B, which is the charging and discharging state of electric energy storage and hydrogen energy storage; C, which is the dispatchable amount of electric energy storage and hydrogen energy storage; SOC, which is the state of charge; HS, which is the hydrogen storage capacity; P, which is the electric power (kW); Q, which is the cooling / heating power (kW); η, which is the efficiency; u, which is the state mark; U, which is the equipment power; σ, which is the energy loss rate of itself; W, which is the energy storage capacity (kWh); t sis the scheduling time interval (h); in the subscripts: BT is the energy storage battery, FC is the hydrogen fuel cell, Chr is charging, dis is discharging, grid is the purchase of electricity from the grid, excess is the sale of electricity to the grid, CL is the controllable load, C is the cooling load power, H is the heating load power, E is the electrical load power, and r is the rated power / capacity. SOC BT HS is the state of charge of the energy storage battery; FC is the hydrogen storage capacity of the hydrogen fuel cell; P BT，chr The power for charging the energy storage battery; P BT，dis The discharge power of the energy storage battery; P grid P is the power purchased from the power grid; excess To sell electricity to the grid; Q FC，chr Cooling / heating power for charging hydrogen fuel cells; Q FC，dis Discharge cooling / heating power for hydrogen fuel cells; Q CL，C is the controllable load cooling load power; Q CL，H is the controllable load heat load power; Q CL，E is the controllable load power; BT is the energy loss rate of the energy storage battery itself; σ FC is the energy loss rate of the hydrogen fuel cell itself; η BT，chr Charging power efficiency of energy storage battery; U BT，chr Power of the electric power equipment for charging the energy storage battery; W BT，r U is the rated power / capacity of the energy storage battery; BT，dis The power of the energy storage battery discharging electric power equipment; η FC，dis is the discharge power efficiency of hydrogen fuel cell; U FC，chr Power of the electric power equipment for charging the hydrogen fuel cell; η FC，chr Electric power efficiency for charging hydrogen fuel cells; U FC，dis is the power of the hydrogen fuel cell discharge electric power device; η FC，dis Discharge electrical power efficiency for hydrogen fuel cells.

[0129] Among them, all system states of the system are measurable, and integration can be introduced to reduce or eliminate static errors, changing the discrete state space equation model to an incremental form:

[0130]

[0131] Among them, the prediction step is p, the control step is m, and m≤p, the control amount remains unchanged outside the control time domain, that is, Δu(k+i)=0, i=m, m+1,…, p-1. This is because the control time domain is smaller than the prediction time domain, and the prediction of the future dynamics of the system requires control input in the entire prediction time domain.

[0132] At the current time k, the measured value is x(k), and Δx(k) can be calculated. This is used as the starting point for predicting the future dynamics of the system. Based on the incremental form of the modeling of the photovoltaic hydrogen storage and charging system equipment, the state from time k+1 to time k+p (actually the state increment) can be predicted as follows:

[0133] Δx(k+1|k)=AΔx(k)+BΔu(k);

[0134] Δx(k+2|k)=AΔx(k+1|k)+BΔu(k+1);

[0135]

[0136] Δx(k+m|k)=AΔx(k+m-1|k)+BΔu(k+m-1);

[0137]

[0138] Δx(k+p|k)=AΔx(k+p-1|k)+BΔu(k+p-1);

[0139] Among them, k+1|k represents the prediction of time k+1 at time k. The controlled output from time k+1 to time k+p can be predicted in the incremental form of the output light hydrogen storage charging system equipment modeling.

[0140] y(k+1|k)=CΔx(k+1|k)+y(k);

[0141] y(k+2|k)=CΔx(k+2|k)+y(k+1|k);

[0142]

[0143] y(k+m|k)=CΔx(k+m|k)+y(k+m-1|k);

[0144]

[0145] y(k+p|k)=CΔx(k+p|k)+y(k+p-1|k);

[0146] The p-step prediction output data and the m-step control input data are as follows:

[0147]

[0148] It can be seen that the prediction equation of the system in the future p steps is: p (k+1|k)=S x Δx(k)+S u ΔU m (k)+S yy(k); where:

[0149] The controlled output reference value Y can be determined r (k+1) and control input reference value U r (k);

[0150] Knowable

[0151]

[0152] The final optimal solution of the control increment ΔU is obtained by solving the approximate method * (k), obtain the first control increment of the control increment sequence The active dispatch instruction with the smallest increment of control quantity is u(k)=u(k-1)+Δu(k).

[0153] In one embodiment, the rolling optimization of the controlled output tracks the active output of the system equipment obtained by the preset light storage hydrogen charging load power prediction model, which can minimize the input control increment and execute the active scheduling instruction with the smallest control increment.

[0154] Embodiment 4

[0155] Figure 6 Schematic diagram of the structure of a system scheduling device provided according to Embodiment 4 of the present invention. Figure 6 As shown, the device includes: a data determination module 61, a power prediction module 62, an output determination module 63 and an instruction generation module 64.

[0156] The data determination module 61 is used to determine the current weather characteristic information and the current system cooling, heating and electric load power;

[0157] The power prediction module 62 is used to determine the cold and hot electric load power at the prediction time based on the preset light hydrogen storage charging load power prediction model, according to the current weather characteristic information and the current system cold and hot electric load power; wherein the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on a self-learning teaching and learning algorithm;

[0158] The output determination module 63 is used to determine the active output of the photovoltaic hydrogen storage charging system equipment based on the cold and hot electric load power and the performance parameters of the photovoltaic hydrogen storage charging system equipment at the prediction time;

[0159] The instruction generation module 64 is used to determine the active scheduling instruction of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction.

[0160] In the embodiment of the present invention, the current weather characteristic information and the current system cold, hot and electric load power are determined by a data determination module. The power prediction module determines the cold, hot and electric load power at the prediction moment based on the preset solar hydrogen storage charging load power prediction model according to the current weather characteristic information and the current system cold, hot and electric load power, so as to achieve accurate prediction of the cold, hot and electric load power at the prediction moment. The output determination module determines the active output of the solar hydrogen storage charging system equipment based on the cold, hot and electric load power at the prediction moment and the performance parameters of the solar hydrogen storage charging system equipment. The instruction generation module determines the active scheduling instruction of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction, thereby achieving determination of the active scheduling instruction by the active output, improving the rationality of the active scheduling instruction, and ensuring the rational distribution of the internal power of the solar hydrogen storage charging system equipment.

[0161] In one embodiment, the system scheduling device further includes:

[0162] A data processing module is used to obtain weather characteristic information of historical days, weather characteristic information of the day to be tested, cold, heat and electricity load power of historical days, and cold, heat and electricity load power of the day to be tested, and perform data preprocessing on the weather characteristic information of historical days and the cold, heat and electricity load power of historical days associated with the weather characteristic information of each historical day;

[0163] A category determination module is used to cluster the weather characteristic information of historical days according to the K-means clustering algorithm and mark the weather category of the weather characteristic information of each historical day;

[0164] A power determination module is used to determine the weather category to which the weather characteristic information of the day to be measured belongs according to the weather characteristic information of the day to be measured as the first category, and extract the cold, heat and electricity load power of the historical day associated with the weather characteristic information of the historical day corresponding to the first category as the cold, heat and electricity load power to be matched;

[0165] The module for determining the acquaintance day is used to determine the power similarity between each to-be-matched cold, heat and electric load power and the cold, heat and electric load power of the to-be-tested day, arrange the power similarities in descending order, and take the to-be-matched cold, heat and electric load power corresponding to the preset number of power similarities arranged in front as the target cold, heat and electric load power of the to-be-tested day, and take the historical day to which the target cold, heat and electric load power belongs as the similar day;

[0166] A historical power determination module, used to determine the cooling, heating and electricity load powers at historical moments with a preset sliding window width before a similar day as the historical cooling, heating and electricity load powers;

[0167] A training set determination module is used to form a training set and a test set according to the weather characteristic information of the day to be tested, the cooling, heating and electricity load power of the day to be tested, the historical cooling, heating and electricity load power, and the target cooling, heating and electricity load power;

[0168] The model training module is used to input the training set into the preset light hydrogen storage charging load power prediction model, determine the optimal weight and the optimal threshold based on the self-learning teaching and learning algorithm until the preset number of iterations is reached, and train the preset light hydrogen storage charging load power prediction model based on the optimal weight and the optimal threshold.

[0169] In one embodiment, the model training module is specifically used to:

[0170] Initialize a preset number of student populations; each population represents a set of weights and thresholds;

[0171] Based on each student population, the training set is input into the preset light storage hydrogen charging load power prediction model to determine the mean square error corresponding to each student population;

[0172] The student population corresponding to the minimum mean square error among the mean square errors is taken as the teacher population, and the corresponding weights and thresholds of each student population are updated according to the teacher population and the preset self-learning weights until the preset number of iterations is reached, and the weights and thresholds corresponding to the teacher population are determined as the optimal weights and optimal thresholds.

[0173] In one embodiment, the system scheduling device further includes:

[0174] The predicted power determination module is used to input the weather characteristic information of the day to be tested and the cold, heat and electricity load power of the day to be tested into the trained preset light storage hydrogen charging load power prediction model to obtain the predicted cold, heat and electricity load power;

[0175] A deviation determination module, used to determine the deviation value between the predicted cooling, heating and electric load power and the target cooling, heating and electric load power;

[0176] The parameter adjustment module is used to adjust the parameters of the preset light hydrogen storage charging load power prediction model according to the deviation value until the deviation value reaches the target value.

[0177] In one embodiment, the data determination module 61 includes:

[0178] A weather determination unit, used to determine weather information, and determine current weather characteristic information according to the weather information; wherein the current weather characteristic information at least includes daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature and daily minimum temperature;

[0179] The power determination unit is used to detect the system cooling, heating and electricity load power at the current moment as the current system cooling, heating and electricity load power.

[0180] In one embodiment, the output determination module 63 includes:

[0181] A parameter determination unit, used to determine the performance parameters of the photovoltaic hydrogen storage and charging system equipment and extract a preset active output confirmation algorithm; wherein the performance parameters at least include the refrigeration coefficient of the photovoltaic hydrogen storage and charging system equipment and the thermal efficiency of the photovoltaic hydrogen storage and charging system equipment;

[0182] The output determination unit is used to determine the active output of the photovoltaic hydrogen storage and charging system equipment according to the preset active output confirmation algorithm, performance parameters and the cold, hot and electric load power at the predicted time.

[0183] In one embodiment, the instruction generation module 64 includes:

[0184] A threshold determination unit, used to extract a pre-configured active output threshold;

[0185] The first instruction generating unit is used to generate a negative active dispatch instruction when it is determined that the active output is greater than a preset active output threshold, so as to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold;

[0186] The second instruction generating unit is used to generate a positive active dispatch instruction when it is determined that the active output is less than a preset active output threshold, so as to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold.

[0187] A system scheduling device provided in an embodiment of the present invention can execute a system scheduling method provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0188] Embodiment 5

[0189] Figure 7 1 is a schematic diagram of the structure of an electronic device for implementing a system scheduling method of an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0190] like Figure 7As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0191] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0192] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a system scheduling method.

[0193] In some embodiments, a system scheduling method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of a system scheduling method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform a system scheduling method in any other appropriate manner (e.g., by means of firmware).

[0194] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0195] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0196] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0197] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0198] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0199] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0200] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0201] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A system scheduling method, characterized in that: include: Determine current weather characteristics and current system cooling, heating and electrical load power; Based on a preset light hydrogen storage charging load power prediction model, the cold, heat and electricity load power at the prediction time is determined according to the current weather characteristic information and the current system cold, heat and electricity load power; wherein the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on a self-learning teaching and learning algorithm; Determine the active output of the solar hydrogen storage and charging system equipment based on the cold and hot electric load power at the predicted time and the performance parameters of the solar hydrogen storage and charging system equipment; An active scheduling instruction of the solar hydrogen storage charging system equipment is determined based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction.

2. The method according to claim 1, characterized in that The initial training process of the preset light hydrogen storage charging load power prediction model includes: Obtain weather characteristic information of historical days, weather characteristic information of the day to be tested, cold, heat and electricity load power of historical days, and cold, heat and electricity load power of the day to be tested, and perform data preprocessing on the weather characteristic information of the historical days and the cold, heat and electricity load power of the historical days associated with the weather characteristic information of each of the historical days; Clustering the weather characteristic information of the historical days according to the K-means clustering algorithm, marking the weather category of the weather characteristic information of each historical day; According to the weather characteristic information of the day to be measured, determine the weather category to which the weather characteristic information of the day to be measured belongs as the first category, and extract the cold, heat and electricity load power of the historical day associated with the weather characteristic information of the historical day corresponding to the first category as the cold, heat and electricity load power to be matched; Determine the power similarity between each of the to-be-matched heat and electricity load powers and the heat and electricity load powers of the to-be-tested day, arrange the power similarities in descending order, and take the heat and electricity load powers to be matched corresponding to the preset number of power similarities arranged in front as the target heat and electricity load powers of the heat and electricity load powers of the to-be-tested day, and take the historical day to which the target heat and electricity load powers belong as the similar day; Determine the cooling, heating and electricity load powers at the historical moments of the preset sliding window width before the similar day as the historical cooling, heating and electricity load powers; A training set and a test set are formed according to the weather characteristic information of the day to be tested, the cooling, heating and electricity load power of the day to be tested, the historical cooling, heating and electricity load power, and the target cooling, heating and electricity load power; The training set is input into the preset photohydrogen storage charging load power prediction model, and the optimal weight and the optimal threshold are determined based on the self-learning teaching and learning algorithm until the preset number of iterations is reached, and the preset photohydrogen storage charging load power prediction model is trained based on the optimal weight and the optimal threshold.

3. The method according to claim 2, characterized in that The inputting the training set into the preset light hydrogen storage charging load power prediction model, determining the optimal weight and the optimal threshold based on the self-learning teaching and learning algorithm until the preset number of iterations is reached, includes: Initialize a preset number of student populations; wherein each of the populations represents a set of weights and thresholds; Based on each student population, the training set is input into the preset light hydrogen storage charging load power prediction model to determine the mean square error corresponding to each student population; The student population corresponding to the mean square error with the smallest value among the mean square errors is taken as the teacher population, and the corresponding weights and thresholds of each student population are updated according to the teacher population and the preset self-learning weights until the preset number of iterations is reached, and the weights and thresholds corresponding to the teacher population are determined as the optimal weights and optimal thresholds.

4. The method according to claim 2, characterized in that: After training the preset light hydrogen storage charging load power prediction model based on the optimal weight and the optimal threshold, the method further includes: The weather characteristic information of the day to be tested and the cold, heat and electricity load power of the day to be tested are input into the trained preset light storage hydrogen charging load power prediction model to obtain the predicted cold, heat and electricity load power; Determining a deviation value between the predicted cooling, heating and electric load power and the target cooling, heating and electric load power; The parameters of the preset light hydrogen storage charging load power prediction model are adjusted according to the deviation value until the deviation value reaches the target value.

5. The method according to claim 1, characterized in that The determining of the current weather characteristic information and the current system cooling, heating and electric load power includes: Determine weather information, and determine current weather characteristic information according to the weather information; wherein the current weather characteristic information at least includes daily average light intensity, daily average temperature, daily average humidity, daily average cloud cover, daily maximum temperature and daily minimum temperature; The system cooling, heating and electricity load powers at the current moment are detected as the current system cooling, heating and electricity load powers.

6. The method according to claim 1, characterized in that The determining the active output of the solar hydrogen storage and charging system equipment based on the cold and hot electric load power at the predicted time and the performance parameters of the solar hydrogen storage and charging system equipment includes: Determine the performance parameters of the solar hydrogen storage and charging system equipment, and extract the preset active output confirmation algorithm; wherein the performance parameters at least include the refrigeration coefficient of the solar hydrogen storage and charging system equipment and the thermal efficiency of the solar hydrogen storage and charging system equipment; The active output of the solar hydrogen storage and charging system equipment is determined according to the preset active output confirmation algorithm, the performance parameters and the cold, hot and electric load powers at the predicted time.

7. The method according to claim 1, characterized in that The determining of the active power dispatching instruction of the solar hydrogen storage charging system device based on the active power output, so that the solar hydrogen storage charging system device controls the system state of the solar hydrogen storage charging system device based on the active power dispatching instruction, comprises: Extract pre-configured active output thresholds; When it is determined that the active output is greater than a preset active output threshold, a negative active dispatch instruction is generated to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold; When it is determined that the active output is less than a preset active output threshold, a forward active scheduling instruction is generated to adjust the system state of the solar hydrogen storage charging system equipment until the active output is equal to the active output threshold.

8. A system scheduling device, characterized in that: include: A data determination module is used to determine the current weather characteristic information and the current system cooling, heating and electric load power; A power prediction module is used to determine the cold, heat and electric load power at the prediction time based on a preset light hydrogen storage charging load power prediction model, according to the current weather characteristic information and the cold, heat and electric load power of the current system; wherein the preset light hydrogen storage charging load power prediction model is a back propagation network model optimized based on a self-learning teaching and learning algorithm; An output determination module, used to determine the active output of the solar hydrogen storage charging system equipment based on the cold and hot electric load power at the predicted time and the performance parameters of the solar hydrogen storage charging system equipment; An instruction generation module is used to determine the active scheduling instruction of the solar hydrogen storage charging system equipment based on the active output, so that the solar hydrogen storage charging system equipment controls the system state of the solar hydrogen storage charging system equipment based on the active scheduling instruction.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the system scheduling method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the system scheduling method according to any one of claims 1 to 7 when executed.

Images