New energy power generation production analog simulation system

Abstract

The invention relates to the technical field of power grids, in particular to a new energy power generation production analog simulation system which comprises a state simulation module, a sequence arrangement module, a condition verification module, a simulation evaluation module and a state interaction module. The method comprises the following steps: performing analog simulation on a production state of a new energy station to determine a simulation sample set; according to the simulation density of the simulation sample set, establishing operation state vectors, and performing time sequence arrangement on the operation state vectors to generate a power generation state sequence; acquiring simulation conditions of a plurality of time intervals according to the time interval of the maximum output scene under the power generation state sequence, and determining a performance index of each simulation condition; verifying the influence degree of each simulation condition, and analyzing the goodness of fit of the performance indexes under each simulation condition to obtain dominant factors of each simulation condition; according to the dominant factors of each simulation condition, constructing a simulation state guide diagram, and determining a simulation control instruction; accuracy and efficiency of new energy power generation production analog simulation are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and specifically to a new energy power generation production simulation and emulation system. Background Art

[0002] The large-scale access of new energy power generation brings challenges to the operation of power grids. Before the grid connection of new energy power stations, testing and evaluating their grid connection performance is an important means to ensure the safe and stable operation of power grids. Limited by capacity and technology, at present, only inverters or power generation units can be tested and field-tested, and the grid connection performance of new energy power stations needs to be solved by means of simulation. When conducting simulation and emulation, the existing technology generally adopts single-factor simulation, ignoring factors such as equipment and grid constraints, which only reduces the matching accuracy between the sample and the actual scenario.

[0003] For example, Chinese Patent Publication No. CN112508338A discloses a comprehensive evaluation system for the grid connection performance of new energy power stations, belonging to the technical field of power grids. The system includes a grid operation data interface for transmitting grid operation data to a complex grid condition simulation module, a new energy power station model parameter identification and verification module for verifying the model using test data, a grid complex condition simulation module for conducting complex grid simulation, a grid regional operation risk determination module for determining operation risks, a new energy power station clustering module for dividing new energy power station groups, a support vector machine module for dynamically modeling the grid connection performance indicators of new energy power stations, an indicator correction module for correcting the grid connection performance evaluation indicators of new energy power stations, a new energy power station comprehensive evaluation module for evaluating the grid-connected new energy power stations, and a new energy power station evaluation result display and output module for displaying and outputting the evaluation results.

[0004] For example, Chinese Patent Publication No. CN113408924A discloses a planning method for a park integrated energy system based on statistical machine learning. The planning method includes: obtaining historical meteorological data, performing power flow calculations on the power generation power of distributed photovoltaic power sources, the power generation power of thermal systems, and the gas supply power of gas systems to obtain the operating state variables of the park integrated energy system, establishing an objective function based on this operating state variable, setting constraint conditions, and iteratively solving the planning problem composed of the objective function and the constraint conditions to obtain a distributed photovoltaic power source siting and sizing planning scheme.

[0005] In the prior art, the performance rankings and evaluations under grid connection are emphasized, and the outdoor temperature is combined to predict the temperature change of the energy system. However, these conditions tend to be set based on manual experience, easily missing low-probability and high-risk scenarios. At the same time, the prior art focuses on the timeliness of indicators rather than dealing with the state transitions and other methods generated by multiple indicators, resulting in the easy neglect of the correlations between multiple factors in a complex grid environment. The accuracy of grid prediction simulation is more affected by single factors, and finally the simulation accuracy is reduced, which is not conducive to predicting the changes in the grid under the influence of multiple factors. Summary of the Invention

[0006] To solve the above technical problems, the technical solution adopted by the present invention is: a new energy power generation production simulation system, including: a state simulation module, a sequence arrangement module, a condition verification module, a simulation evaluation module, and a state interaction module; wherein, the output end of the state simulation module is connected to the sequence arrangement module, the output end of the sequence arrangement module is connected to the condition verification module, the output end of the condition verification module is connected to the simulation evaluation module, and the output end of the simulation evaluation module is connected to the state interaction module.

[0007] The state simulation module is used to simulate and simulate the production state of new energy power stations with the input data of each new energy power station and the simulation density of energy output, and determine the simulation sample set.

[0008] The sequence arrangement module is used to establish the operation state vectors of each simulation operation state according to the simulation density of the simulation sample set, and perform time series arrangement in the form of the output of the operation state vectors to generate a power generation state sequence.

[0009] The condition verification module is used to obtain the simulation conditions within multiple time intervals according to the time interval of the maximum output scenario in the power generation state sequence; and determine the performance indicators of each simulation condition based on the energy output density and power generation state of the power generation state sequence.

[0010] The simulation evaluation module is used to verify the influence degree of each simulation condition on the power generation situation, analyze the goodness of fit of the performance indicators under each simulation condition, and obtain the dominant factors of each simulation condition.

[0011] The beneficial effects of the present invention are as follows: First, the present invention generates a simulation sample set by combining multi-source inputs such as meteorological conditions, equipment status, and historical data. By using the calculation method of the output probability sequence for the simulation sample set, after calculating its probability and correlating it with the simulation density, the collected data are coupled to improve the authenticity of the simulation samples during generation. Then, for the formed operation state vectors, according to their state indexes, power truncation constraints, and change rate constraints, the forms such as power generation states are restricted to reduce the data with power mutations in the simulation sample set, so as to ensure that the equipment of each new energy power station can maintain a safe operation state during the advancement of the time series.

[0012] 2. After setting the parameters that generate the maximum output power within the time interval of the maximum output scenario in the power generation state sequence, the present invention processes the simulation conditions in combination with their manifestation forms in different time intervals, and describes the data of the power generation state sequence with indicators such as the average output power and the curtailment rate, so as to generate performance indicators that meet each simulation condition to adapt to the parameters under different process combinations.

[0013] 3. The present invention clusters the simulation conditions according to single and multiple combinations to determine the influence degree between the simulation conditions under various clusterings, and uses transfer entropy and the like to represent different fitting indicators in the single-classification cluster and the multi-classification cluster, so as to illustrate the manifestation forms of the data of each cluster after the simulation conditions are clustered. Then, by identifying the relatively overlapping parts of the data in the cluster, the factors that mainly cause the simulation conditions are obtained, and then circular combinations are carried out in the way of node orientation of each factor to explain the control instructions required by the current simulation conditions in the scenario from single factor to multi-factor, so as to realize the closed-loop from state analysis to real-time decision-making, assist the later maintenance personnel to timely maintain its data simulation, and improve the environmental adaptability of the simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below in conjunction with the drawings and embodiments.

[0015] Figure 1 It is a system schematic diagram of a new energy power generation production simulation system.

[0016] Figure 2 It is a process schematic diagram of the state simulation module of a new energy power generation production simulation system.

[0017] Figure 3 It is a process schematic diagram of the sequence arrangement module of a new energy power generation production simulation system.

[0018] Figure 4 It is a process schematic diagram of the condition verification module of a new energy power generation production simulation system.

[0019] Figure 5 It is a process schematic diagram of the simulation evaluation module of a new energy power generation production simulation system.

[0020] Figure 6 It is a process schematic diagram of the state interaction module of a new energy power generation production simulation system. DETAILED DESCRIPTION OF THE INVENTION

[0021] Embodiments of the present invention will be described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as a limitation of the present invention. For those without specific technical or conditions noted in the embodiments, they shall be carried out according to the technologies or conditions described in the literature in this field or according to the product manuals.

[0022] Referring to Figure 1 , a new energy power generation production simulation system, comprising: a state simulation module, a sequence arrangement module, a condition verification module, a simulation evaluation module, and a state interaction module; wherein, the output end of the state simulation module is connected to the sequence arrangement module, the output end of the sequence arrangement module is connected to the condition verification module, the output end of the condition verification module is connected to the simulation evaluation module, and the output end of the simulation evaluation module is connected to the state interaction module.

[0023] The state simulation module is used to simulate and simulate the production state of the new energy power station with the input data of each new energy power station and the simulation density of the energy output, and determine the simulation sample set.

[0024] The sequence arrangement module is used to establish the operation state vectors of each simulation operation state according to the simulation density of the simulation sample set, perform time series arrangement in the output form of the operation state vectors, and generate the power generation state sequence.

[0025] The condition verification module is used to obtain the simulation conditions within multiple time intervals according to the time interval of the maximum output scenario in the power generation state sequence; and determine the performance indicators of each simulation condition based on the energy output density and power generation state of the power generation state sequence.

[0026] The simulation evaluation module is used to verify the influence degree of each simulation condition on the power generation situation, analyze the goodness of fit of the performance indicators under each simulation condition, and obtain the dominant factors of each simulation condition.

[0027] The state interaction module is used to construct a simulation state guidance diagram according to the dominant factors of each simulation condition, and determine the simulation control instructions according to the guidance routes of each simulation condition in the simulation state guidance diagram.

[0028] Preferably, the input data of each new energy power station includes but is not limited to the power generation energy of the current new energy power station, historical meteorological data, historical output data, equipment status, and the theoretical output power of the new energy power station. The meteorological data is used to illustrate the environment when the current new energy power station conducts new energy power generation, and the equipment status is used to illustrate whether the equipment in the new energy power station can work normally. At the same time, the equipment status will illustrate the equipment parameters of the power generation equipment in the new energy power station to determine its theoretical output power, as well as the relative distribution of historical data to identify whether the current new energy power station can work normally.

[0029] Preferably, the simulation density of the energy output uses the historical output data of the output power of the new energy power station, and adopts the kernel density estimation method to generate its output probability distribution. Then, after extracting the indexes from the output probability distribution, it is used as the simulation density of the energy output at this time.

[0030] Preferably, as Figure 2 shown, the implementation method of the state simulation module includes: combining the input data of each new energy power station using meteorological conditions, equipment status, and grid conditions respectively to generate simulation samples with different combined structures, and recording the data proportion under each combined structure; the combined structure represents combining the historical data and other contents of the new energy power station using meteorological conditions, equipment status, and grid conditions respectively, explaining the output power size under different conditions, so as to determine the current new energy power generation situation.

[0031] Calculate the probability distribution of the input data of each new energy power station to generate an output probability distribution sequence; according to the output probability distribution sequence, extract the simulation density under the energy output.

[0032] Construct the mapping relationship between the simulation density and the simulation samples with different combined structures, and the mapping relationship between the output probability distribution sequence and the simulation samples. Based on the double-layer mapping relationship, generate a simulation sample set.

[0033] For example, taking the scenario with wind power generation and photovoltaic power generation as an example, the meteorological condition combination will be represented as a combination in three dimensions of wind speed stratification, light intensity stratification, and temperature stratification. Classify the output power of the new energy power station under these data combinations to find out the situation with the highest power generation efficiency and output power.

[0034] Among them, the wind speed stratification can be divided into three gears according to the IEC standard: Cut-in wind speed, with a wind speed of 3 - 4 m / s; rated wind speed, with a wind speed of 12 - 15 m / s; Cut-out wind speed, with a wind speed of 25 m / s. Distinguish and identify the wind speed conditions in the wind power generation scenario, and then extract the corresponding output power.

[0035] The light intensity stratification is divided into three gears according to the characteristics of photovoltaic power generation: low light, medium light, and strong light. Use the values of 500 W / m² and 800 W / m² as the data for dividing the three gears to identify its light power generation situation. When only one of photovoltaic power generation and wind power generation exists in the new energy power station, it is divided only by the existing one.

[0036] The temperature stratification is divided according to the ambient temperature or the working temperature of the equipment, with its average value as the division method. Set multiple intervals according to the percentage of the working temperature deviating from the average value in the historical data, such as more than 10%, more than 30%, and more than 50% divided into multiple numerical intervals.

[0037] The device status combinations link the actual operating conditions of the device with the output current and power in the form of normal operation, derated operation, and faulty operation, and are divided into multiple combined structures.

[0038] The grid condition combinations are combined according to the content of voltage fluctuation, frequency deviation, and fault type. The voltage fluctuation is divided into forms of ±5% rated voltage and ±10% rated voltage. The frequency deviation represents the situations at 49.5Hz, 50Hz, and 50.5Hz. The fault type can illustrate situations such as three-phase short circuit, single-phase grounding, and voltage drop to 0.2pu, etc., to describe the power generation forms of new energy power stations.

[0039] Preferably, the probability distribution calculation of the input data of each new energy power station can adopt the form of Gaussian distribution. After calculating the probability values of the output power of the new energy power station, the input data corresponding to each probability value are formed into an output power probability distribution sequence according to the probability value selection, and then the index values of the output power probability distribution sequence are extracted as the simulation density of the energy output; such as the peak value, variance, average value, and minimum output power of the output power probability distribution sequence, to be used as its output simulation density.

[0040] Preferably, the double-layer mapping relationship is mapped according to the values of the simulation density in the simulation samples of different combined structures, and according to the probability values of the output power probability distribution sequence and the probability of the corresponding simulation samples existing within the input time; it illustrates the distribution mapping of the simulation density generated by the combination of different combined structures, associates the simulation density with the proportion and combination method of the combined structure, and then maps the probability value corresponding to the simulation density with the time and other forms associated with the input data; to achieve a mapping method of structure-density decoupling and probability-behavior coupling, to illustrate the complete simulation link from static configuration to dynamic response; the first mapping ensures the combination coverage of the samples, and the second mapping injects the time-series probability characteristics, and finally generates a simulation data set with both structural diversity and output authenticity.

[0041] In an embodiment of the present invention, the simulation operation state refers to the comprehensive operation performance of new energy power stations such as wind farms and photovoltaic power stations at specific time points or under specific conditions during the simulation process. It reflects the actual output characteristics of the system under given input conditions such as wind speed, light intensity, temperature, etc. and operation constraints such as equipment capacity, grid dispatching requirements, etc. In wind power simulation, a simulation operation state may be a wind speed of 8m / s, a fan speed of 12rpm, an output power of 1.5MW, and a grid frequency of 50Hz, while another simulation operation state may be a wind speed of 15m / s and an output power of 2.0MW.

[0042] The operating state vector combines multiple sets of data corresponding to power generation at the simulation density. For example, the average value and standard deviation corresponding to the simulation density are used as a label, and then the active power, reactive power, wind speed, light intensity, grid frequency, timestamp, and equipment status of each data in this set of data are described, indicating the power that can be output and the corresponding situation at the simulation density at a specific time point, as shown in Table 1.

[0043] Table 1. Schematic of the operating state vector

[0044]

[0045] Table 1 shows the dimensions included in the operating state vector. These dimensions represent the main parameters involved in the power generation production of new energy power stations. Based on these parameters, it can be evaluated whether the current new energy power station can produce electricity normally. These data are obtained from the input data, and the operating state vector is arranged in the form of a time series to obtain a power generation state sequence to simulate the continuous changes in actual operation.

[0046] As Figure 3 shown, the implementation method of the sequence arrangement module also includes: based on the index values corresponding to the simulation density, establishing a state index between each new energy power station. The state index represents the active power, reactive power, and equipment operation status in each energy power station, as well as the parts directly used for new energy power generation such as wind speed and light intensity in the environment. An index of the data of these new energy power stations is established to facilitate the subsequent simulation and processing of the operating state.

[0047] When a state index query is triggered, according to the data quantity and operating state vector corresponding to each state index, the operating state of the target equipment corresponding to the state index is indicated; the target equipment represents the equipment for power generation in the new energy power station.

[0048] Determine the output form of the operating state vector according to the operating state of the target device. The output forms are generally divided into numerical type, boolean type, and enumeration type; the numerical type directly represents the numerical form of the operating state vector. For example, [wind speed = 8.2 m / s, fan speed = 15 rpm, gearbox temperature = 65 °C, power generation = 2.5 MW]. In this form, the specific values of the operating state vector are described, and then the power generation state sequence is formed in the form of a time series. The boolean type indicates whether the target device is normal, such as indicating whether the device is in a normal state in the form of "yes / no" or "1 / 0", which is used for subsequent warning or logical judgment of simulation conditions. The enumeration type describes the mode or category of the device operating state through tags, such as [operating mode = grid-connected power generation, cooling state = automatic, wind direction deviation = 5° to the left]. These can describe the power generation situation when the devices in the new energy power station generate electricity, and identify whether the states of each device change during the power generation of new energy by the changes of these noted tags in the sampling at consecutive time points.

[0049] Preferably, when the operating state vector is established, the data in its simulation sample set will be standardized, and after normalizing each data, a dimensionless power generation state sequence for subsequent processing is formed.

[0050] Preferably, when arranging the time series in the output form of the operating state vector, it is also necessary to process the output power fluctuation limit and limit the difference between the predicted power and the actual power to solve the problems of suppressing the output power fluctuation and optimizing the curtailment rate in the power generation scenario; at this time, the physical constraints of each new energy power station are used to adjust its power generation state sequence, and the specific situation where the output power value can be under the minimum fluctuation is made. At this time, based on the operating state vector of the new energy power station in historical data, it is processed to obtain a power generation state sequence composed of a time interval, an operating state vector, and an output power confidence interval. The output power confidence interval refers to the confidence interval corresponding to the output power.

[0051] Preferably, the implementation method of arranging the time series in the output form of the operating state vector includes: setting power truncation constraints and rate of change constraints for the output power at adjacent times in the operating state vector in sequence; the power truncation constraint means that the output power shall not exceed the maximum output of the device. For example, if the rated power of the fan is 2.5 MW, then the output power in the simulation sample set shall not exceed this value; the rate of change constraint means the constraint of the power change amount at adjacent times with its maximum ramp rate, that is, Among them, represents the power change amount at adjacent times, represents the maximum ramp rate, and the maximum ramp rate is the maximum change rate at which new energy power generation equipment, such as wind turbines, photovoltaic power stations, etc., can safely adjust their output power per unit time; represents the time period difference at adjacent times.

[0052] According to the status index, the operating state vector is introduced into the double circular queue of power cut-off constraint and rate of change constraint. When the double circular queue points to any status index, the data corresponding to the status index is combined into a power generation status sequence in the form of a time series. The pointing here indicates that the operating state vector satisfies the power cut-off constraint and the rate of change constraint. At this time, the generated simulation data is more biased towards the real environment. By dynamically adjusting the predicted value in this way, it can be ensured that the rate of change of the output power of the new energy power generation equipment is always within the safe range, thus ensuring the stable operation of the power grid and the safe operation of the equipment.

[0053] Preferably, the double circular queue represents a queue for retrieving the current operating state vector with power cut-off constraint and rate of change constraint. When all the data in the queue satisfy their constraints, the corresponding operating state vectors are arranged into an output power generation status sequence.

[0054] In an embodiment of the present invention, the maximum output scenario refers to the scenario with the maximum output power that new energy power generation can reach under specific simulation conditions. The simulation conditions are a set of parameter combinations designed around the maximum output form, which are used to cover the full operating conditions range of the system. Among them, the maximum output scenario will represent the data corresponding to the maximum output power obtained within a time interval of the power generation status sequence. The parameters of these data are extracted into simulation conditions. At the same time, the power generation status sequence will contain multiple time intervals to identify the relevant states of the simulation conditions under what combination of simulation conditions, so as to select the performance indicators for subsequent processing of the simulation conditions.

[0055] Such as Figure 4 shown, the implementation method of the condition verification module further includes: predicting multiple potential simulation conditions for the current time interval according to the simulation conditions of the previous time interval.

[0056] Merge the potential simulation conditions of the current time interval with the simulation conditions to obtain the target simulation conditions of the current time interval.

[0057] Statistically analyze the occurrence probability of the target simulation conditions. If the occurrence probability of the target simulation conditions is greater than the probability threshold, the corresponding target simulation conditions are used as the output simulation conditions.

[0058] Preferably, when predicting potential simulation conditions, a LSTM time series prediction model is used for prediction; the LSTM time series prediction model uses the simulation conditions of the previous time region as input to determine which information is input into the current time interval. For example, through a fixed-length window, such as the window length = 6 time steps, corresponding to 1-hour data to intercept the historical segment as the model input. Each window represents a training sample, and the label is the simulation condition of the next time step at the end of the window. When the model starts, the hidden state and cell state are initialized, usually set to zero vectors, indicating no prior knowledge.

[0059] After processing step by step in time, the forget gate will decide which historical information to discard based on the current input, such as the wind speed at the previous time step and the previous hidden state. For example, if the current wind speed suddenly increases, the forget gate may discard the old memories related to low wind speed.

[0060] The input gate will filter the data to determine which new information, such as the current wind speed and temperature change, to store in the cell state. For example, the input gate may focus on recording wind speed mutation events and weaken the data with steady changes.

[0061] The cell state will fuse the forgotten old information and the new information to generate the current cell state, which is stored as long-term memory. The output gate generates the current hidden state based on the cell state and the current input, which is used as the input for the next time step.

[0062] Finally, the LSTM traverses the entire historical sequence, gradually updates the hidden state and the cell state, and finally generates a context vector that synthesizes the historical pattern; the cell state of the LSTM is input into the fully connected layer and mapped to the potential simulation conditions for the current time interval. The fully connected layer may output the probabilities of the three states of high, medium, and low wind speed as 0.6, 0.3, and 0.1. Then, the Softmax function is used to ensure the normalization of the output probabilities, reflecting the relative likelihood of each potential condition.

[0063] Because the current simulation conditions represent the data measured by simulating all the contents in the operation state vector, the LSTM will output a data containing the output power and various scenarios. Then, a probability value is set for the numerical value of the simulation conditions, and this probability value is calculated in the form of a Gaussian distribution, such as simulating conditions of combinations like "high wind speed (probability 0.6), moderate temperature (probability 0.7), equipment healthy (probability 0.85)". As for the LSTM, it is the content recorded in the prior art, so there will be no more explanations here.

[0064] Preferably, when merging the potential simulation conditions for the current time interval with the simulation conditions, it is actually taking the Cartesian product of the predicted potential simulation conditions and the existing simulation conditions to generate a data set containing all possible combinations; each combination represents a simulation scenario, covering information such as the equipment environment, and this data set is regarded as the target simulation conditions.

[0065] After that, the occurrence probability is calculated for the target simulation conditions. This occurrence probability is obtained by solving the conditional probabilities of multiple simulation conditions within the corresponding combination of the target simulation conditions. Then, the probability threshold will select this value according to the simulation conditions in the combination of the target simulation conditions. For example, the average value of the occurrence probabilities of the corresponding simulation conditions in the maximum output power scenario is used to set this probability threshold. It is also possible to select a threshold that can cover historical scenarios as much as possible. For example, the occurrence probabilities of historical simulation conditions are sorted, and the minimum value of the occurrence probability when the cumulative probability reaches 90% is used as the probability for screening the current target simulation conditions, so as to screen the simulation conditions that can be as close as possible to the historical scenarios.

[0066] Preferably, taking wind power generation and photovoltaic power generation as examples, the energy output density can represent the power generation amount of a unit area of photovoltaic panels within a certain period of time, or the actual power generation amount of a unit area of wind turbine swept area within a certain period of time. This certain period of time is in units of days or years, to illustrate the power generation amount of new energy power generation under the unit area measurement method under the simulation conditions corresponding to the power generation state sequence.

[0067] The power generation state represents the sum of the standard deviations of the output power of this power generation state sequence and the curtailment rate. The curtailment rate represents the proportion of unused renewable energy caused by conservative prediction or equipment limitations. This part represents the ratio of the unused available power existing under the current simulation conditions to the output power, used to illustrate the underutilized situation existing in the power generation state sequence. The curtailment rate is the ratio value calculated by dividing the difference obtained by subtracting the output power in the current sequence from the theoretical power by the theoretical power. Then, the values corresponding to the energy output density and the power generation state are used as the performance indicators of their simulation conditions.

[0068] In an embodiment of the present invention, when verifying the influence degree of each simulation condition on the power generation situation, first, the simulation conditions are divided into simulation conditions under multiple similar scenarios, then the mapping relationship between each simulation condition and the performance indicator after fitting is statistically analyzed to find out how the simulation condition affects the output power of its power generation. Finally, the influence degree and main factors among the simulation conditions are output to describe the influence on the simulation index in the power generation environment; for example, taking the simulation condition as the input condition, clustering correction is performed to evaluate the fitting form of the performance indicator and the concurrent form of the performance indicator of each simulation condition under the simulation simulation, so as to determine the influence of each simulation condition on the simulation process.

[0069] Such as Figure 5As shown in the figure, the implementation method of the simulation evaluation module further includes: taking each simulation condition as an input, determining the single-classification cluster of a single simulation condition and the multi-classification cluster of combinations of multiple simulation conditions; when setting the classification clusters, clustering each simulation condition in the way of DBSCAN, classifying the content of a single simulation condition and combinations of multiple simulation conditions, to obtain a set of multiple similar scenarios, such as dividing out actual scenario contents representing power generation such as "high wind speed and low irradiance". Among them, a single simulation condition represents a set of parameters that reach the maximum output power within a time region, to illustrate what similar scenarios exist in the scenario corresponding to this set of parameters, facilitating subsequent analysis of the conditions that lead to the maximum output power in this scenario; the set of combinations of multiple simulation conditions is more inclined to illustrate how to transition to the conditions for the maximum output power in the next time interval when the maximum output power is reached in a time interval, and is used to illustrate whether there are corresponding causal associations under multiple scenarios, resulting in state transitions and other contents of the existence of multiple simulation conditions. That is, the single-classification cluster is a set obtained by clustering starting from the set of parameters corresponding to a single simulation condition, to obtain simulation conditions similar to this single simulation condition; the multi-classification cluster is a set found to be similar to the scenario of the combination of multiple simulation conditions after combining multiple simulation conditions. For example, in a certain case, the single-classification cluster can be a wind speed parameter cluster, and the multi-classification cluster can be a wind speed-illumination joint cluster, thus illustrating the influence of the obtained simulation conditions on the output power.

[0070] According to the obtained multi-classification clusters and single-classification clusters, set the influence degree of each simulation condition based on the value of the output power of each simulation condition within the multi-classification cluster and the single-classification cluster.

[0071] Preferably, the influence degree of each simulation condition can be represented based on the ratio of the mean value of the output power of the current simulation condition in the single-classification cluster to the standard deviation within the corresponding multi-classification cluster, to represent the influence degree of this simulation condition. The higher the ratio, the more concentrated the output power is under this condition or the smaller the deviation from the overall standard deviation, indicating better stability or less influence by this condition; on the contrary, it indicates that this condition has a greater influence on the output power, resulting in an increase in data dispersion. Then, by comparing the influence degrees corresponding to the corresponding simulation conditions, it can be known which set of parameters has a more significant influence, and the part with a low ratio may represent the key sensitive points of its system performance, to illustrate whether the current equipment's power generation is stable.

[0072] As shown in Table 2, compare the content corresponding to the current single-classification cluster with the corresponding data of the multi-classification cluster, to determine the relationship between the data corresponding to each parameter in the combined form of each simulation condition under different scenarios, and obtain the goodness of fit corresponding to this simulation condition.

[0073] Table 2. Schematic Diagram of Classification Clusters

[0074]

[0075] If there is a sample i1 in the single-class cluster of the current simulation conditions, the silhouette coefficient within the single-class cluster can be expressed as :[[]] ; where represents the average distance from sample i1 to other samples in the current single-class cluster, which is calculated based on the performance metrics of the simulation conditions. When calculating the performance metrics, since the performance metrics will be expressed in the form of standard deviations of power generation, curtailment rate, and output power, at this time, after normalizing the samples of the parameter set represented by the simulation conditions in vector form with other samples within the current single-class cluster, the Euclidean distance calculation method is used to obtain its distance value, that is, let ; where represents the number of samples in the current single-class cluster, and the values of i1 and j1 range from 1 to n1; and respectively represent the normalized values of the energy output density of sample i1 and sample j1, and respectively represent the normalized values of the standard deviations of the output power of sample i1 and sample j1, and respectively represent the normalized values of the curtailment rate of sample i1 and sample j1. represents the average distance from sample i1 to samples in the nearest single-class cluster, and the calculation method of this distance value is the same as the above calculation of distance, so it will not be described here; the larger the value, the higher the compactness and the better the separation of the single-class cluster combined by a single simulation condition; it is used to illustrate what kind of association exists between the output powers represented by these performance metrics after using simulation condition clustering.

[0076] The conditional probability of its multi-class cluster can be calculated based on the formula of conditional probability, taking the currently sampled simulation data as the basis.

[0077] The transfer entropy represents the transfer situation from the parameters corresponding to one simulation condition to the parameters corresponding to another simulation condition. At this time, based on the state of multiple simulation condition combinations transferring to the state of another group of multiple simulation condition combinations the transfer entropy ; that is, at this time the transfer entropy ; where represents the conditional probability of state and state , represents state The probability, which can be calculated based on the form of the Gaussian distribution, represents the relationship between two multi-class clusters with the conditional probability of the performance metric of the parameter set after combining multiple simulation conditions changing from one value state to another. As for the value ranges of i2 and j2, they are set according to the number of multi-class clusters where state transitions occur.

[0078] After completing the goodness-of-fit processing of its performance metric, based on the goodness-of-fit, the dominant factors of its simulation conditions are identified. That is, the implementation method of the simulation evaluation module also includes: separately judging the goodness-of-fit of the performance metrics corresponding to each simulation condition under single-class clusters and multi-class clusters, sorting the influence degrees and goodness-of-fit of each simulation condition, and using the sorted output result as the dominant factor of each simulation condition.

[0079] Preferably, when performing feature sorting, for a single-class cluster, if there is a simulation condition whose influence degree and goodness-of-fit can take the highest value, then this content with the highest value can be used as the dominant factor of this simulation condition. For example, the profile coefficient of wind speed is the highest and is the single dominant factor affecting the output power of current power generation; if the transfer entropy of wind speed plus grid frequency is the highest in a multi-class cluster, then the content of the combination of multiple simulation conditions is used as its dominant factor at this time; therefore, it is necessary to perform multi-objective processing on the content that can be output by single-class clusters and multi-class clusters to obtain a set of dominant factors most relevant to the current simulation conditions under multiple output dominant factors, which is used as the guidance for describing the state conversion of the power system later, so as to analyze the form of the change trend of the output power of the current new energy station under the conditions of the currently set data.

[0080] The implementation method of sorting the influence degrees and goodness-of-fit of each simulation condition includes: for the influence degrees of simulation conditions within single-class clusters and multi-class clusters, obtain the duration, output power, and parameter range of each simulation condition within single-class clusters and multi-class clusters, and determine the key features and associated simulation conditions of the current simulation condition.

[0081] For each key feature, identify the overlapping data within each time interval to form an overlapping data set.

[0082] Compare the overlapping data set with the associated simulation conditions, and select the data parameter with the largest goodness-of-fit value after comparison as the output dominant factor.

[0083] Preferably, the key features of the simulation conditions are extracted based on the duration, output power, and parameter range values within single-class clusters and multi-class clusters. For example, the key features are the data corresponding to the simulation conditions with the highest average output power or the longest duration selected from the single-class cluster; the associated simulation conditions are the data combinations of multiple simulation conditions with the highest transition probability or the highest output power in the multi-class cluster.

[0084] After that, the data points with duplicates of the key features in each time region are extracted. These data represent the general situation of the equipment power generation in the new energy power station during power generation. Then, the coincidence data set is compared with the associated simulation conditions to capture the synergistic effects of single factors and multi-factors. After that, the same parts are processed when comparing the key features with the associated simulation conditions, and the goodness of fit calculated for the simulation conditions corresponding to the same parts is selected. Since the goodness of fit of the simulation conditions is divided into multi-simulation condition combinations and single simulation conditions, the recognized goodness of fit will output the part corresponding to the maximum value selected from the two to illustrate the data mainly affected at this time. For example, in one case, the goodness of fit calculated for the wind speed interval in the single-class cluster is 0.85, and the goodness of fit calculated for the wind speed interval plus the light intensity interval in the multi-class cluster is 0.92. At this time, the parameter interval corresponding to the multi-class cluster will be selected as the leading factor for subsequent simulation judgment to illustrate the main factors mainly affecting new energy power generation in the current situation.

[0085] In an embodiment of the present invention, as Figure 6 shown, the implementation manner of the state interaction module further includes: using each leading factor as a node and the influence degree of the simulation conditions corresponding to the leading factor as the weight of the edge between the nodes to form a simulation state guidance graph.

[0086] According to the goodness of fit when each node in the simulation state guidance graph is connected, set the guidance of the simulation state guidance graph, and use the shortest connection path in each guidance as the guidance route of each simulation condition.

[0087] Set a threshold trigger strategy according to the output power on the guidance route of each simulation condition, and select a simulation control instruction according to the threshold trigger strategy.

[0088] Preferably, the threshold trigger strategy is set based on the average value of the data corresponding to each simulation condition in the historical data, or set using the average value of the time interval closest to the currently processed simulation condition; for example, when the output power of the corresponding node on the guidance route is greater than or less than the threshold, different simulation control instructions are selected.

[0089] Preferably, the orientation of the simulation state orientation diagram is determined based on the transfer entropy calculated in the multi-class cluster under the simulation conditions. If the value of the transfer entropy from the simulation conditions corresponding to the current dominant factor to the simulation conditions of another dominant factor is greater than zero, the orientation is considered to be from the current dominant factor to the other dominant factor; otherwise, the direction is opposite.

[0090] Preferably, the shortest connection path in each orientation is determined by the shortest path algorithm, with the values corresponding to each dominant factor plus the weights of each node, to obtain the shortest path corresponding to each node after the nodes are connected by the orientation.

[0091] Preferably, the implementation method of selecting the simulation control instruction according to the threshold trigger strategy further includes: for the orientation route of each simulation condition, extracting the sub-paths of each orientation route, and superimposing the output power of each sub-path with the output power of the corresponding orientation path to obtain the path power of each orientation route.

[0092] Perform threshold retrieval according to the path power of each orientation route to identify the threshold trigger strategy of the current orientation route; when the path power of the orientation route is greater than the threshold of the threshold trigger strategy, select the simulation control instruction according to the minimum difference in the output power of the orientation path in the adjacent time interval.

[0093] When the path power of the orientation route is less than the threshold of the threshold trigger strategy, select the simulation control instruction according to the simulation condition corresponding to the maximum output power in the current orientation route.

[0094] The sub-paths of the above-mentioned orientation routes are obtained by dividing the orientation route of each simulation condition into multiple indivisible sub-paths, summing the output power on the sub-paths to obtain the path power of each orientation route, and then searching for the simulation control instruction set in the database according to the value of the path power. When exceeding the threshold, the form of minimizing power fluctuation will be selected for processing at this time, that is, selecting the control instruction corresponding to the minimum difference in the database to avoid equipment overload or grid instability caused by power mutation. If it is lower than the threshold of its threshold trigger strategy, it is necessary to quickly increase the path power, for example, giving priority to processing the node corresponding to the highest output power to quickly restore its power generation efficiency.

[0095] For example, the simulation control instruction for the minimum difference selection is a simulation control instruction that slowly reduces the current path power to avoid overloading devices such as wind turbines that generate electricity; while the simulation control instruction corresponding to the maximum output power is to adjust the parameters of the device operation to increase the overall output power, etc., facilitating subsequent interference by maintenance personnel, etc. with the new energy power generation situation. At the same time, after the simulation control instruction is output, the state simulation module is carried out again with the data regulated by the simulation control instruction to implement the logical control method from simulation → verification → simulation, so as to be able to perform multiple simulation adjustments of its production simulation to improve the accuracy of the data output by the simulation.

[0096] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention, and still be covered by the protection scope of the present invention.

Claims

1. A new energy power generation production simulation system, characterized in that, Including: A state simulation module, which is used to simulate and model the production state of new energy power stations with the input data and simulation density of energy output of each new energy power station, and determine a simulation sample set; A sequence arrangement module, which is used to establish an operation state vector for each simulation operation state according to the simulation density of the simulation sample set, perform time series arrangement in the output form of the operation state vector, and generate a power generation state sequence; A condition verification module, which is used to obtain simulation conditions within multiple time intervals according to the time interval of the maximum output scenario in the power generation state sequence; determine the performance indicators of each simulation condition based on the energy output density and power generation state of the power generation state sequence; A simulation evaluation module, which is used to verify the influence degree of each simulation condition on the power generation situation, analyze the goodness of fit of the performance indicators under each simulation condition, and obtain the dominant factors of each simulation condition; A state interaction module, which is used to construct a simulation state guidance diagram according to the dominant factors of each simulation condition, and determine a simulation control instruction according to the guidance route of each simulation condition in the simulation state guidance diagram.

2. The new energy power generation production simulation system according to claim 1, wherein The input data of the new energy power station includes but is not limited to the power generation energy of the current new energy power station, historical meteorological data, historical output data, equipment status, and the theoretical output power of the new energy power station.

3. A new energy power generation production simulation system according to claim 1, characterized in that, The implementation method of the state simulation module includes: Combining the input data of each new energy power station with meteorological conditions, equipment status, and grid conditions respectively to generate simulation samples with different combined structures, and recording the data proportion under each combined structure; Performing probability distribution calculation on the input data of each new energy power station to generate an output probability distribution sequence; extracting the simulation density under the energy output according to the output probability distribution sequence; Constructing a mapping relationship between the simulation density and the simulation samples with different combined structures, and a mapping relationship between the output probability distribution sequence and the simulation samples, and generating a simulation sample set based on the double-layer mapping relationship.

4. A new energy power generation production simulation system according to claim 1, characterized in that The implementation method of the sequence arrangement module further includes: Establishing a state index between each new energy power station based on the index value corresponding to the simulation density; When triggering a state index query, indicating the operation state of the target device corresponding to the state index according to the data quantity and operation state vector corresponding to each state index; Determining the output form of the operation state vector according to the operation state of the target device.

5. A new energy power generation production simulation system according to claim 4, characterized in that, The implementation method of performing time series arrangement in the output form of the operation state vector includes: Successively setting power truncation constraints and change rate constraints on the output power at adjacent moments in the operation state vector; Introducing the operation state vector into a double circular queue of power truncation constraints and change rate constraints according to the state index. When the double circular queue points to any state index, combining the data corresponding to the state index in the form of a time series into a power generation state sequence.

6. A new energy power generation production simulation system according to claim 1, characterized in that The implementation method of the condition verification module further includes: Predicting multiple potential simulation conditions for the current time interval according to the simulation conditions of the previous time interval; Combining the potential simulation conditions of the current time interval with the simulation conditions to obtain the target simulation conditions for the current time interval; Statistical analysis of the occurrence probability of the target simulation conditions. If the occurrence probability of the target simulation conditions is greater than the probability threshold, the corresponding target simulation conditions are used as the output simulation conditions.

7. The new energy power generation production simulation system according to claim 1, characterized in that: The implementation methods of the simulation evaluation module also include: Taking each simulation condition as input, determining the single-classification clusters of individual simulation conditions and the multi-classification clusters of combinations of multiple simulation conditions; According to the obtained multi-classification clusters and single-classification clusters, setting the influence degree of each simulation condition based on the values of the output power of each simulation condition within the multi-classification clusters and single-classification clusters; Respectively judging the goodness of fit of the performance indicators corresponding to each simulation condition under the single-classification clusters and multi-classification clusters, sorting the influence degree and goodness of fit of each simulation condition by features, and taking the sorted output result as the leading factor of each simulation condition.

8. A new energy power generation production simulation system according to claim 7, characterized in that The implementation method of sorting the influence degree and goodness of fit of each simulation condition by features includes: For the influence degree of simulation conditions within the single-classification clusters and multi-classification clusters, obtaining the duration, output power, and parameter range of each simulation condition within the single-classification clusters and multi-classification clusters, and determining the key features and associated simulation conditions of the current simulation condition; For each key feature, identifying the overlapping data within each time interval to form an overlapping data set; Comparing the overlapping data set with the associated simulation conditions, and selecting the data parameter with the largest goodness of fit value after comparison as the leading factor of the output.

9. A new energy power generation production simulation system according to claim 1, characterized in that The implementation methods of the state interaction module also include: Taking each leading factor as a node and the influence degree of the simulation condition corresponding to the leading factor as the weight of the edge between the nodes to form a simulation state orientation graph; Setting the orientation of the simulation state orientation graph according to the goodness of fit when the nodes in the simulation state orientation graph are connected, and taking the shortest connection path in each orientation as the orientation route of each simulation condition; Setting a threshold trigger strategy according to the output power on the orientation route of each simulation condition, and selecting a simulation control instruction according to the threshold trigger strategy.

10. A new energy power generation production simulation system according to claim 9, characterized in that, The implementation method of selecting a simulation control instruction according to the threshold trigger strategy also includes: For the orientation route of each simulation condition, extracting the sub-paths of each orientation route, and superimposing the output power of each sub-path with the output power of the corresponding orientation path to obtain the path power of each orientation route; Performing threshold retrieval according to the path power of each orientation route to identify the threshold trigger strategy of the current orientation route; when the path power of the orientation route is greater than the threshold of the threshold trigger strategy, selecting a simulation control instruction according to the minimum difference in the output power of the orientation path within the adjacent time interval; When the path power of the orientation route is less than the threshold of the threshold trigger strategy, selecting a simulation control instruction according to the simulation condition corresponding to the maximum output power in the current orientation route.

Images