A smart city heating system and heating control method

The smart city heating system addresses inefficiencies in urban heating by using data collection, prediction modeling, and dynamic control to accurately forecast and adjust heat energy distribution, enhancing efficiency and comfort while reducing costs.

CN118856407BActive Publication Date: 2024-12-06DALIAN WORMAN TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411337632.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2024-12-06
Estimated Expiration
2044-09-25

AI Technical Summary

Technical Problem

Existing urban heating systems face inefficiencies due to inaccurate heat demand predictions, leading to excessive heating, energy waste, and insufficient heating, which affects user comfort and health, especially in response to climate change and changes in user behavior patterns.

Method used

A smart city heating system that uses a data acquisition module to collect historical and real-time data, a prediction module to establish a model based on time series analysis and machine learning for accurate demand forecasting, and a control module to dynamically adjust heat energy distribution according to predicted demands.

Benefits of technology

The system enhances energy efficiency by minimizing waste and ensuring user comfort by accurately predicting and responding to heat demands, reducing costs and improving the sustainability of urban heating systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118856407B_ABST
    Figure CN118856407B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent city heating system and a heating control method, wherein the system comprises a data acquisition module for acquiring historical heat consumption data and current environmental parameters of each user; a prediction module for establishing a prediction model based on the historical heat consumption data and the current environmental parameters to predict the future heat demand of each user; and a control module for dynamically adjusting the heat energy distribution of each user according to the predicted future heat demand. The intelligent city heating system and the heating control method solve the problem that the operating efficiency of the heating system is often low due to inaccurate heat demand prediction in the prior art, resulting in two extreme situations: one is excessive heating, resulting in heat energy waste and increased energy costs; the other is insufficient heating, which cannot meet the actual needs of users, thereby affecting the user's comfort and even having a negative impact on health and quality of life.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of urban heating, and in particular to an intelligent urban heating system and a heating control method. Background Art

[0002] In the current urban heating system, with the acceleration of urbanization and the enhancement of environmental protection awareness, the traditional heating model has gradually exposed many shortcomings, especially in terms of energy efficiency and user experience. The emergence of smart city heating methods has brought new solutions for urban district heating. Its core lies in the integration of advanced Internet of Things (IoT), big data analysis and artificial intelligence (AI) technologies, aiming to improve the efficiency and sustainability of the heating system through accurate heat demand prediction and dynamic regulation.

[0003] Existing urban heating systems usually rely on experience or simple historical data to predict users' heat demand. However, this method has great uncertainty and limitations, especially in responding to climate change, changes in user behavior patterns and differences in building energy consumption. Inaccurate heat demand predictions often lead to inefficient operation of the heating system, resulting in two extreme situations: one is overheating, which leads to heat waste and increased energy costs; the other is insufficient heating, which cannot meet users' actual needs, thus affecting users' comfort and even having a negative impact on health and quality of life. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent city heating system and a heating control method to solve the large uncertainties and limitations of the existing technology, especially in response to climate change, changes in user behavior patterns and differences in building energy consumption. Due to inaccurate heat demand prediction, the operation efficiency of the heating system is often low, resulting in two extreme situations: one is excessive heating, resulting in heat waste and increased energy costs; the other is insufficient heating, which cannot meet the actual needs of users, thereby affecting the user's comfort and even having a negative impact on health and quality of life.

[0005] To achieve the above object, the present invention provides the following technical solution: a smart city heating system, the system comprising:

[0006] A data acquisition module is used to obtain historical heat consumption data and current environmental parameters of each user;

[0007] The prediction module is used to establish a prediction model based on historical heat consumption data and current environmental parameters to predict the future heat demand of each user;

[0008] A control module is used to dynamically adjust the heat energy allocation to each user according to the predicted future heat demand;

[0009] The data acquisition module acquires the historical heat consumption data and current environmental parameters of each user including:

[0010] Receive and record the daily heat consumption of each user in the past week from the user end;

[0011] Extract real-time weather data corresponding to each user's location;

[0012] According to the user's location information, the relationship table between the environmental parameters and the user's heat consumption in the same period of history is called;

[0013] Integrate the above historical and real-time data into the system database;

[0014] Use IoT technology to regularly check and update the status of each sensor to ensure the accuracy of the acquired data;

[0015] Among them, the use of Internet of Things technology to regularly check and update the status of each sensor to ensure the accuracy of the acquired data includes maintaining connection with various sensors through Internet of Things technology, regularly sending instructions to check the status of the sensor, and using the sensor health status detection model to monitor the working status of the sensor. The specific formula is: ,

[0016] in, Indicates the sensor at time health status at that time, Indicates time, Indicates the health status of the sensor battery. Indicates the sensor at time The network connection status at the time Indicates the sensor at time The accuracy of data collection at the time Represents a function;

[0017] The prediction module establishes a prediction model based on historical heat consumption data and current environmental parameters to predict the future heat demand of each user including:

[0018] Collect the user's daily heat consumption for at least one year in the past as a historical data sample;

[0019] Combine the user's geographical location and climate conditions to obtain temperature data for the corresponding date;

[0020] The collected historical data samples and weather data are processed using the time series analysis method to train the correlation between the user's individual heat consumption and meteorological factors. The specific formula is: ,

[0021] in, Indicates that the user Daily heat consumption, Indicates the day number. Indicates The ambient temperature of the day, represents the lag period of the time series, is the error term, Represents a function, which is a model based on time series analysis;

[0022] Based on the above correlation results, the expected heating demand for each user in the next period is predicted. Based on the established time series model, the future temperature is predicted and the corresponding heat demand is calculated. The specific formula is: ,

[0023] in, Indicates Daily user heating demand, The predicted The ambient temperature of the day, represents the lag period of the time series, is the serial number of the day, , , , …, are the coefficients in the model trained based on historical data;

[0024] The predicted number of days is days, we can get a set of future heating demands: ,

[0025] in, is the number of days to be predicted;

[0026] The control module dynamically adjusts the heat energy distribution of each user according to the predicted future heat demand, including:

[0027] Check whether each submodule within the system is online and dispatchable to respond to heating requests;

[0028] The instructions are transmitted from the data center to each heat exchange station through network communication facilities to guide the operation;

[0029] Dynamically open and adjust the pumping pressure in the supply pipeline and the valve opening to adjust the output temperature to meet the user's customized demand curve;

[0030] Immediately make capacity increase or conservation adjustment strategies for abnormal situations that exceed the normal working load limit or are under-predicted;

[0031] Implementing remote monitoring ensures that heat supply control can continue to be effective without human intervention.

[0032] Preferably, the processing using a time series analysis method includes:

[0033] Filter and remove extremely abnormal historical daily heat consumption data points to smooth the time series trend;

[0034] Analyze the daily consumption and average temperature change patterns to form an initial prediction formula;

[0035] Based on the temperature difference change rate correction formula, the possible fluctuation range of the next day's temperature is calculated by [(tomorrow's predicted minimum temperature - today's measured minimum temperature) / today's measured minimum temperature], and the consumption factor in the formula is adjusted;

[0036] If the fluctuation exceeds 5%, a correction value is applied to the forecast to reduce the impact of uncertainty.

[0037] Preferably, the correction formula based on the temperature difference change rate includes:

[0038] Determine the daily minimum temperature to predict the weather conditions for the next day;

[0039] When the temperature difference change rate is negative, indicating cooling, the positive correlation coefficient is adjusted on the model input parameter to enhance the accuracy of estimated heating in cold weather;

[0040] When the temperature difference change rate is non-negative, which means that the temperature is rising or remaining the same, the correlation coefficient is kept unchanged or adjusted downward appropriately by no more than 0.01;

[0041] Future heat consumption forecasts are regenerated based on this rule, ensuring that changes in the impact of rapid short-term weather transitions on actual consumption are adapted.

[0042] Preferably, when the temperature difference change rate is a negative number, it specifically refers to:

[0043] Before actual operation, first check the validity and timely upload of temperature sensor data to avoid misjudgment due to equipment failure;

[0044] Setting a warning line such as [tomorrow's expected low temperature minus the measured low temperature on the day] < [1 degree] is one of the trigger points for the significant cold air invasion warning. If this condition is verified, a specific response instruction will be executed;

[0045] Immediately after receiving the warning, the product part of the heat output target quota is increased according to the pre-programmed program;

[0046] The action of increasing the heating supply is terminated based on real-time temperature feedback adjustment until the temperature difference deviates from the estimated range or the same decline does not occur for three consecutive observation intervals.

[0047] Preferably, when the real-time temperature is lower or higher than the predicted value by less than 5 degrees Celsius, the original heating scheme is maintained unchanged:

[0048] Read the actual environmental measurements fed back by the automatic monitoring system, including the current air temperature information and its speed level updated to the minute;

[0049] The set comparison algorithm is used to estimate the temperature difference between the current time value and the last generated plan to obtain the numerical value of the deviation amplitude, that is, the actual measured average temperature - the absolute value of the expected level calculated by the system;

[0050] If the deviation is within the range of [−0.5℃, +0.5℃], it is considered to be within the acceptable control range.

[0051] After the final judgment result indicates that no adjustment measures are needed, the scheduled heating service process will continue to be implemented and the above inspection process will be repeated at a fixed frequency to ensure the continuous tracking of the impact of external natural changes on the heating intensity. The fact-based support system operates effectively.

[0052] A smart city heating control method, using the smart city heating system, is characterized in that the method comprises:

[0053] S1: Obtain historical heat consumption data and current environmental parameters of each user;

[0054] S2: establishing a prediction model based on the historical heat consumption data and current environmental parameters to predict the future heat demand of each user;

[0055] S3: Dynamically adjust the thermal energy distribution of each user according to the predicted future thermal demand.

[0056] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0057] The smart city heating system and heating control method obtain the historical heat consumption data and current environmental parameters of each user through a data acquisition module; the prediction module establishes a prediction model based on the historical heat consumption data and current environmental parameters to predict the future heat demand of each user; the control module dynamically adjusts the heat energy distribution of each user according to the predicted future heat demand, which solves the large uncertainty and limitations of the existing technology, especially in response to climate change, changes in user behavior patterns and differences in building energy consumption. Inaccurate heat demand prediction often leads to low operating efficiency of the heating system, resulting in two extreme situations: one is excessive heating, resulting in heat energy waste and increased energy costs; the other is insufficient heating, which cannot meet the actual needs of users, thereby affecting the user's comfort and even having a negative impact on health and quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the system module of the present invention;

[0059] Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0061] like Figure 1 As shown, a smart city heating system, the system comprises:

[0062] The data acquisition module collects heat demand data from various residential buildings or structures in the city and records the previous heat usage data of these users and their related environmental indicators; such as external factors such as temperature, climate and wind speed. The main purpose of this step is to obtain a large amount of reliable historical heat usage data to lay the foundation for future analysis of heat demand trends and service improvements, and also to better understand the uniqueness and demand preferences of each user or group, such as the preferred heating temperature setting of residents in a certain apartment building.

[0063] The forecasting module uses advanced data analysis technology and intelligent algorithms to build a model based on detailed historical heat consumption data and real-time climate conditions to make a scientific estimate of heat energy consumption in the next period. This stage combines historical heat consumption and existing environmental information to form an estimate of future heating requirements. By using cutting-edge technologies such as machine learning, we can find hidden patterns in past data, allowing us to predict future heating requirements and avoid energy waste and insufficient supply.

[0064] The final control module is the key step to implement flexible heating regulation, that is, to formulate and implement a dynamic heat supply strategy based on the results of the first two stages and the expected future heating needs of various users. This measure can ensure that accurate services are provided according to the actual expectations of users instead of a one-size-fits-all approach. For example, when the cold winter is approaching, the system can find that a certain community may have a large increase in heating demand in the future based on previous analysis. With this prediction, the heating system will automatically increase the flow in the community's heating network in advance. This method of adjusting supply in time according to changes in needs can achieve more efficient use and optimize the allocation of resources, avoiding cost increases caused by excess and dissatisfaction caused by shortages.

[0065] For a more concrete demonstration, here is a specific example. A smart city in the north is in the coldest days of the year. Based on the winter heating situation accumulated over the past decade and the latest weather forecast, the analysis tool concluded that there will be a greater increase in warmth around the Spring Festival this year than usual. The system therefore pre-arranged to increase the transmission of heat energy to each household heating equipment during this period; in addition, it was detected that there were several office buildings near a commercial center in the city that were sparsely populated at night and did not need to operate at full capacity for a long time, so at night, the heat source release in this area was appropriately reduced to reduce unnecessary losses. All of the above are to save costs without affecting the user experience. Through this series of measures, the heating department effectively reduced operating expenses while ensuring the daily heating requirements of the people, and improved the energy efficiency of the entire smart city while meeting comfort. In summary, the use of intelligent methods to predict and respond to the heating and cooling needs of citizens in the actual operation process can not only prevent supply and demand but also avoid cost pressure caused by overcapacity.

[0066] Next, the acquisition module is described to obtain the historical heat consumption data and current environmental parameters of each user. First, the daily heat consumption data of the user in the previous seven days is filtered out from the information received from the user end, and recorded and stored; this data provides the recent demand pattern of each user and lays a solid foundation for predicting the next weather adjustment. For example, a smart heating system can connect to the heat meter of each household, collect the average heat consumption per day, and store this information through the cloud platform.

[0067] After that, the real-time weather conditions corresponding to the current geographical location of each household are obtained from the environmental monitoring network, including but not limited to temperature, humidity and other key climate parameters; such environmental data can help determine the specific heating needs of residents in the short term in the future. For example, in winter, if the real-time temperature in a certain area drops sharply, the system will immediately collect environmental data near the residents in the area and further calculate the adjustment range of heating, thereby improving comfort and reducing the probability of excessive resource consumption.

[0068] Next, the weather characteristics and actual heating usage in similar periods in the past are compared according to the location of the residence or facility. This helps to estimate the trend of future consumption demand changes, making the heating plan more forward-looking and effective. Take a hotel's heating in winter as an example. At the same time (such as the Spring Festival holiday), the preset temperature is adjusted to take into account the possible increase in passenger flow, and the heater operating frequency is adjusted in advance based on the experience of the same time period in the past few years to meet the needs of guests.

[0069] The actual values ​​obtained in the early stage and the historical records of their corresponding geographical locations are then aggregated and transmitted to a centralized data storage facility to ensure data security and easy access; this approach makes data analysis more convenient. For example, all the changes in personal home and surrounding conditions collected above are synchronously recorded in a secure data center, which is convenient for subsequent algorithm models to read and analyze and make optimal regulation recommendations based on this.

[0070] Finally, IoT networking technology is used to regularly check the operating status of each deployed monitoring device to ensure that the measuring tools can detect abnormalities in a timely manner and take corresponding maintenance or reset measures to avoid data failure, which ensures the authenticity, accuracy and reliability of information when making decisions. Assuming that there is a fault in the meter or a problem with data transmission, the IoT system will immediately alert maintenance personnel to locate the faulty node and quickly eliminate interference factors, ensuring that the overall intelligent control network operates efficiently without interruption.

[0071] Among them, the use of Internet of Things technology to regularly check and update the status of each sensor to ensure the accuracy of the acquired data includes maintaining connection with various sensors through Internet of Things technology, regularly sending instructions to check the status of the sensor, and using the sensor health status detection model to monitor the working status of the sensor. The specific formula is: ,

[0072] in, Indicates the sensor at time health status at that time, Indicates time, Indicates the health status of the sensor battery. Indicates the sensor at time The network connection status at the time Indicates the sensor at time The accuracy of data collection at the time Represents a function.

[0073] Through the Internet of Things technology, the system will regularly maintain connections with various sensors and send status detection instructions to them. Based on the feedback information obtained from the sensors, the system calculates the current health status of the sensors. Key parameters include the health of the sensor battery, the network connection status, and the accuracy of data collection. These parameters are calculated through the function Analysis is performed to determine the overall health status of the sensor. This ensures that the sensor is always in good condition during actual work, thereby improving the accuracy of data collection and the overall performance of the system. In this way, the system can detect any abnormal status of the sensor in a timely manner and take corresponding measures based on the detection results, such as replacing batteries, adjusting network connections, or correcting data collection modes, thereby ensuring the long-term stable operation of the system. In addition, regular monitoring of sensor status can effectively reduce maintenance costs, avoid data collection errors caused by sensor failures, and ensure that all types of data are accurate during transmission.

[0074] In another implementation, some parameters in the sensor health status detection model can be replaced according to the actual application scenario. For example, if the sensor type is different, environmental factors such as temperature and humidity can be added as variables in the model to more comprehensively monitor the health status of the sensor. In addition, the function It can also be adjusted as needed to adapt to more complex application requirements by using different algorithms (such as machine learning algorithms or statistical analysis methods).

[0075] Next, the prediction module is described to establish a prediction model based on historical heat consumption data and current environmental parameters to predict the future heat demand of each user.

[0076] Collect the user's daily heat consumption for at least one year in the past as a historical data sample;

[0077] Combine the user's geographical location and climate conditions to obtain temperature data for the corresponding date;

[0078] The collected historical data samples and weather data are processed using the time series analysis method to train the correlation between the user's individual heat consumption and meteorological factors. The specific formula is: ,

[0079] in, Indicates that the user Daily heat consumption, Indicates the day number. Indicates The ambient temperature of the day, represents the lag period of the time series, is the error term, Represents a function, which is a model based on time series analysis;

[0080] Based on the above correlation results, the expected heating demand for each user in the next period is predicted. Based on the established time series model, the future temperature is predicted and the corresponding heat demand is calculated. The specific formula is: ,

[0081] in, Indicates Daily user heating demand, The predicted The ambient temperature of the day, represents the lag period of the time series, is the serial number of the day, , , , …, are the coefficients in the model trained based on historical data;

[0082] The predicted number of days is days, we can get a set of future heating demands: ,

[0083] in, The number of days to predict.

[0084] First, the daily heat consumption data of the user for at least one year in the past is collected as a historical data sample, and the ambient temperature data of the corresponding date is obtained in combination with the climatic conditions of the user's geographical location. The historical data and meteorological data are processed using the time series analysis method to train a correlation model between the user's heat consumption and meteorological factors. The model estimates the user's heat consumption demand under different meteorological conditions through regression analysis. Based on this model, the system can predict the future temperature and use the time series model to calculate the user's heating demand in a specific time period in the future. Through the implementation of the present invention, the heating demand of each user can be accurately predicted, avoiding the waste of resources or user discomfort caused by excessive or insufficient heating in the traditional heating system. This method can not only improve the heating efficiency, but also effectively reduce the heating cost and energy consumption. At the same time, the prediction model takes into account the user's geographical location and local climatic conditions, so it is more personalized and accurate, and can flexibly adapt to different user needs. In addition, since the model is based on time series analysis, it can capture the regularity in historical data, making the heating prediction more reliable.

[0085] In another embodiment, in addition to using the time series analysis method, other machine learning algorithms (such as random forests, neural networks, etc.) can also be used to build a correlation model between user heat consumption and meteorological factors. Different algorithms may perform better when dealing with nonlinear relationships, thereby improving the accuracy of the prediction. In addition, the selected meteorological parameters can also be expanded. In addition to temperature, environmental factors that affect heating demand, such as humidity and wind speed, can also be included to further optimize the effect of the prediction model.

[0086] First, one or more models suitable for analyzing and predicting changes in heat demand in urban heating systems need to be selected based on a specific type of machine learning algorithm. This step is about deciding which or which forecasting models are best suited for analyzing heat consumption, usually based on a range of candidate algorithms (such as regression trees, neural networks, etc.). For example, in an urban environment, time series decomposition or long short-term memory (LSTM) networks may be chosen to estimate demand because of the presence of clear daily or weekly patterns.

[0087] Secondly, by combining the past heat consumption information and the recent weather trend data to set the initial parameter values ​​or weights, a basic forecasting model that can reflect the actual situation is initially established, and the necessary parameters are adjusted for these initial values ​​in order to improve the effect of the algorithm. What is mentioned here is to set the initial weights based on the existing data records and recent climate conditions, and use this method to fine-tune the basic framework of the selected forecasting algorithm. If the historical heat consumption of a location shows a difference between weekends and weekdays, this difference should be taken into account when initializing the model, and the expected value change curve between Friday and Monday should be adjusted accordingly.

[0088] Then, after the model is established and initially runs for a period of time, it is necessary to promptly incorporate the heat energy consumption of the last few days, and combine the current actual temperature and other meteorological phenomena as feedback loop input to continue training and adjusting parameters to improve the accuracy of predictions and make them more in line with the latest developments. This link refers to combining the latest actual usage data collected by the system with the events occurring under the current external conditions, repeatedly using the established model to learn new patterns to better adapt to the changing situation. When a site encounters a sudden cold wave that causes residents to frequently turn on their heaters in their homes, this new information should be immediately included in the calculation range to ensure that the next action recommendation has a sufficient and reliable support basis.

[0089] Next, based on the output of the training results from the previous process, we can get the data on how much heat energy users in different spatial ranges (i.e. various subdistricts or community units) are expected to consume in the next stage (calculated as one hour at the shortest) until the whole week. The work at this stage is mainly to apply the results of the previous efforts, so as to lay the foundation for the municipal agencies to arrange reasonable energy allocation and formulate other related plans, and help improve resource efficiency and reduce unnecessary waste. If the forecast indicates that a residential block may face a significant increase in the pressure of heater startup time in the next 24 hours, measures must be taken immediately to ensure that enough energy is supplied to avoid any possibility of a supply crisis.

[0090] Finally, a variety of statistical analysis tools (such as T-value test or confidence interval test) are used to comprehensively evaluate whether the selected model's ability to respond correctly to various abnormal climate conditions is reliable, and to confirm that there is no systematic bias to ensure overall stability and controllability. The task of this stage is to ensure that various technical means are used to thoroughly prove that the formulated strategy can not only work well under normal circumstances, but also has the ability to handle those rare but quite impactful situations. When the local area experienced several consecutive rounds of abnormally heavy rainfall during the summer of a certain year, resulting in the overall outdoor temperature being much lower than the same period of the previous year, the accuracy of the conclusions drawn from the above process must be carefully checked to ensure that all residents receive adequate protection and the quality of service remains unchanged.

[0091] Next, the control module is described to dynamically adjust the heat distribution of each user according to the predicted future heat demand. First, it includes checking whether each submodule within the system is online and dispatchable so that it can respond to the heating request; this means checking all relevant hardware devices such as pumps, heating units, controllers, etc. to ensure their functionality and connectivity before the system is ready to serve users, so that these components can receive central instructions at any time; assuming that a large number of heating demands are received on a cold morning, the control system will automatically verify the status of each submodule to ensure that all heat exchange stations, heaters and other equipment are available and can respond in time.

[0092] The next step described is to forward control instructions from the centralized data center to the heating devices in each area through the network communication architecture to command the execution of specific actions; this means that the operation commands issued by the data center are sent to the on-site control units of each site through pre-deployed dedicated data lines to guide their behaviors such as starting or stopping heating; for example, on a cloudy or windy day, it is necessary to increase or decrease the heat supply to certain areas. After the data center receives the updated information of the demand, it sends adjustment information to the corresponding heat exchange station through a stable communication link, such as increasing the air supply flow, increasing the heating level and other operation signals.

[0093] The third step involves dynamically changing the pump thrust in the supply line and changing the heat delivered by adjusting the valve opening and closing to meet the customer's customized service model; this procedure ensures consistency between the output of the supply side and the demand of the user side and achieves the most efficient utilization effect; for example, in order to achieve the goal of comfortable warmth without wasting electricity, after detecting that the room temperature in the user's home is lower than the preset standard, the control system will automatically increase the pipeline pressure, open a larger hot water circulation flow and lower the inlet opening of the cooling point to gradually increase the temperature until it returns to the desired range.

[0094] Subsequently, timely measures are taken to avoid system overload or save energy consumption to prevent losses when the load exceeds the design standard or unexpected problems occur. This stage includes starting additional auxiliary equipment or reducing the scale of operation to cope with sudden changes when it is monitored that the utilization efficiency of current resources is close to or has reached the limit. For example, if an extreme weather event occurs on the same day, causing a surge in immediate heating requests from all users, the system may immediately activate reserve equipment to bear part of the load or reduce heating in other non-essential places to allocate more energy to ensure the needs of important areas.

[0095] The final step is to implement all-round real-time observation to ensure that the automatic operation system can continuously and effectively control the entire process of heat distribution without relying on human control. This measure emphasizes that even when there is no operator on duty, the system must ensure that normal operation is not disrupted. For example, after the staff on duty leave late at night, it is still necessary to formulate logical judgment conditions in advance so that the automation device can adjust its operation according to changes in actual conditions and continue to complete the specified functional goals, thereby maintaining the stability of the heating level.

[0096] Next, the specific steps of establishing a prediction model based on the historical heat consumption data and current environmental parameters to predict the future heat demand of each user are described: the daily heat consumption of users in the past at least one year is collected as a historical data sample. This step refers to the systematic collection of records of daily heat energy use by each user in a specific time period. This long-term historical data analysis can better understand the heating habits and energy consumption patterns of each user or user group. In some cases, such as the annual consumption data of all residents in a community, it will help us build a more universal heating behavior model to predict future heat loads.

[0097] The next step is to combine the user's geographical location with the climate conditions to obtain the temperature data of the corresponding date. This step aims to collect the historical conditions of the relevant temperature, humidity and other climate change factors in the area where each household is located and match them with the user's daily heating consumption in the same time period. This association allows the model to identify the specific extent to which external environmental changes such as temperature and weather conditions affect the demand for thermal energy. In an example, this can be achieved by integrating the geographical location of residents in a community in a northern city and the historical data of winter temperature changes.

[0098] The third step is to use time series analysis technology to process data. In this process, the results of the first two steps need to be used to train the algorithm model to find the internal connection or dependence between the energy consumption of each household and various weather factors; for example, through ARIMA, SARIMA or long short-term memory artificial neural network models to capture the potential relationship between daily heat load and seasonal cycle fluctuations. This step is to input the collected historical information and climate change data into appropriate data mining tools. For example, in the given example, it means analyzing the actual heating consumption of each heating cycle of the above residential group and the daily weather forecast information, so as to obtain the specific function model expression of each household's heat consumption intensity being constrained by weather change factors.

[0099] The final stage is to deduce the expected heating calorific value required by each consumer in a certain period in the future based on the results of the various parameters obtained above, and generate a final demand forecast result file to be fed back to the heating company as an auxiliary judgment basis for adjusting the heating intensity and the operating frequency of the equipment; this stage uses the relationship equations obtained from the second stage of training to infer the daily supply required by users in the next heating cycle, and transmits these estimated results to the municipal heating company or the building property management department to facilitate the formulation of subsequent operational plans and decisions. For example, in this case, it means using the heating load forecasting plan that has been developed for the target household group to predict the amount of heating energy that each household will need to consume in the next few months, thereby ensuring that the heating service for residents' lives is carried out smoothly, reliably, efficiently and environmentally friendly.

[0100] Next, the specific process of using time series analysis methods is described. First, a comprehensive review of the historical daily heat consumption data set is conducted to remove possible outliers or significant deviations. Such preprocessing steps are necessary and effective, which can prevent extreme values ​​from interfering with or distorting the accuracy of subsequent forecasts and ensure that the time series is more stable and coherent. Then, based on the processed data, a statistical relationship is established to explore the relationship between daily heat consumption and local outdoor air temperature. This phase involves pairing the heating consumption data from the past few days or even longer with the corresponding temperature records, and discovering some patterns or mathematical representations as the basis for subsequent forecasting work; next, considering the unpredictable factors of weather that will directly affect future heat consumption, we also need to fine-tune our calculation equations based on temperature forecast information, especially in the case of expected sharp temperature changes. The temperature difference here refers to the ratio of the predicted low temperature of tomorrow minus the measured minimum temperature of the current day, and the resulting temperature rise and fall ratio will be used as the key basis for adjusting the consumption coefficient; the last step is to set conditional judgment: if the aforementioned temperature change percentage is higher than the predetermined risk threshold of 5%, in order to reduce the model prediction error, it is necessary to introduce a compensation mechanism in the formal estimation to compensate for the negative impact and uncertainty caused by sudden meteorological events. For example, in the case of a cold front that is expected to cause a sharp drop in temperature after a series of warm days, this mechanism can ensure that the final heat consumption estimate for the next day reflects the actual situation as much as possible, rather than generating a large forecast distortion based on outdated trends.

[0101] Next, the specific steps based on the temperature difference change rate correction formula are described. First, confirming the daily temperature records, especially the lowest temperature, is a critical step. These temperature data can provide an initial estimate of the weather changes of the next day and directly affect the judgment and adjustment of heating demand. Especially in winter, such data can predict the signal of the coming of cold air and prepare sufficient heating resources in advance. Then, when it is determined that the temperature difference change rate is negative, that is, when the temperature begins to drop, it is necessary to increase the value of a correlation coefficient called 03 in the model. By increasing the value of this parameter, it can be ensured that the system can better adjust the heating forecast for upcoming or ongoing cooling events. The direct effect of this is that it can more accurately predict that users may need more heat supply in the coming days and make corresponding adjustments.

[0102] Next, if the calculated temperature difference change rate is positive or zero, that is, when the temperature is rising or there is no obvious change, the coefficient should either remain unchanged or be lowered, but the decrease should not exceed 0.01. Because even if the temperature warms up or remains unchanged, it does not mean that the control accuracy of the heating system can be relaxed. This consideration of fine adjustment is because even on warm days, there may be a phenomenon that some users still require a higher heating level due to environmental factors.

[0103] Based on the results after the above-mentioned corrective measures, we will have an updated data version for estimating future heat energy consumption. The advantage of doing this is that the model can adjust itself in a timely manner according to the temperature fluctuations in a short period to ensure that the final expected actual consumption quantity suitable for the current climate environment for heating is obtained. For example, if the night temperature is suddenly detected to drop from 5°C the previous night to 2°C at a monitoring station in a certain area, the above-mentioned 03 correlation coefficient can be adjusted upward according to this temperature difference value to cope with the greater heating load that may be brought by the low temperature during the following daytime period. And when the night temperature rises to about 8 degrees Celsius the next day, this factor will return to the initial set value to match the heat consumption expectation under the new temperature conditions. By repeating this cycle, the precision service supply level of heating in the smart city can be improved.

[0104] Next, describe what the negative temperature difference change rate specifically refers to: The first step before actual operation is to check the effective and timely upload of the temperature sensor data to avoid misjudgment due to equipment failure. In this process, the primary task of the system is to ensure that the obtained temperature difference information is reliable and error-free. This includes whether the collected data comes from effective temperature sensors and whether these data have been accurately and quickly uploaded to the control system. Through such verification, any wrong decisions caused by hardware problems in the processing logic can be effectively avoided.

[0105] Set a warning line, that is, the expected low temperature tomorrow minus the measured low temperature today < 1 degree; 1 degree is regarded as the second warning condition for significant cold air intrusion. After confirming that the current temperature change trend is indeed turning towards a significant drop in temperature, the pre-set alarm mechanism will be activated; the logic here is to determine whether it is necessary to take actions to prevent the adverse effects brought by extreme cooling by comparing the known forecast and the latest measured temperature information.

[0106] After detecting the temperature drop and confirming that the drop has reached the set warning standard, the system will automatically increase the planned heat supply of the heating system according to the pre-written control software. The amount of increased heating intensity mentioned here should be an increment equivalent to 0.3 (03) times the预定 heating value to cope with the impact of the upcoming round of cold high-pressure activities on the ambient temperature and prepare appropriate energy supplement in advance.

[0107] Once the intervention action of increasing the energy supply of the city's heat pipeline is started, the latest weather conditions in the surrounding areas will be monitored in real time, and the temperature readings obtained from actual observations will be continuously fine-tuned to adapt to the changes in external climate conditions until the monitored outdoor temperature fluctuations do not match the previous forecast range or at least the falling trend of the temperature level described by the set warning line is not observed in the recent statistical cycles. Only then can it be decided to resume normal supply and stop increasing the amount. This dynamic adjustment method can more accurately and effectively control energy supply to match the actual user heating needs, reduce unnecessary resource waste and improve the heating service level.

[0108] Next, it is described that if the real-time temperature is lower or higher than the predicted value by less than 5 degrees Celsius, the original heating plan will remain unchanged: The first paragraph first mentions reading the actual environmental measurement values ​​fed back by the automatic monitoring system. This operation means that the system must collect real-time data from a series of meteorological detection stations or other environmental sensing elements deployed in the periphery. This not only includes an accurate quantitative description of the temperature in the air, but also involves tracking and observing the frequency of its changes. This information can usually help managers understand the fluctuations in external conditions that need to be faced during the operation of urban heating in a timely manner. For example, the system recorded that the outdoor temperature was 10°C at 2 o'clock in the afternoon of that day.

[0109] The next stage involves comparing the distance between the latest actual observations collected and the estimated distance based on the model preset based on historical data or weather reports through a predefined mathematical tool to obtain the so-called temperature difference amplitude value. The algorithm used here should be a simple and fast calculation logic. For example, the forecast result of the last time predicted that the temperature of the air at this moment would be 11℃, so based on the comparison between the two, it can be calculated that the absolute value difference is as much as 1℃.

[0110] After determining the size of the above price difference, it is necessary to test whether it is within the allowable deviation threshold. This step is actually to determine under what error range the current action can be considered appropriate and effective. Only when the detected change index does not exceed the set range of ±0.5 degrees can it be considered within the control error boundary. Assuming that the difference in this example is exactly 1 degree Celsius, because it is greater than the specified standard 0.5℃ upper limit, the system will not default to not adjusting the strategy and continue the original plan, but may enter the processing procedure of rearranging the heating intensity or frequency.

[0111] The last step is to implement subsequent operations according to the decision conclusions after the previous inspection. When it is known that there is no need to change any existing work arrangements and there is no need to increase the heat supply, the established policy will be maintained. However, at the same time, it is necessary to ensure that rounds of monitoring and evaluation are carried out regularly to confirm the impact of the actual phenomenon of continuous changes on the intensity of heat supply. If the system starts the above overall process every thirty minutes, assuming that the temperature measurement result at 2 pm is within the standard allowable range, it will repeat the above four-step cycle at a subsequent time point half an hour later to ensure the adaptability, accuracy and safety of the overall operation mechanism. Therefore, the system will resume monitoring, comparison and judgment at 2:30 a.m. half an hour later to determine whether the temperature change has an impact on the established plan, and continue to implement the corresponding strategy.

[0112] like Figure 2 As shown, the present invention also provides a smart city heating method, the method comprising:

[0113] S1: Obtain historical heat consumption data and current environmental parameters of each user;

[0114] S2: establishing a prediction model based on the historical heat consumption data and current environmental parameters to predict the future heat demand of each user;

[0115] S3: Dynamically adjust the thermal energy distribution of each user according to the predicted future thermal demand.

[0116] First, the historical heat consumption data of each user and the real-time environmental parameters, such as outdoor temperature and humidity, are obtained from the database of the urban heating system. Then, these historical data and environmental parameters are combined through artificial intelligence algorithms to establish a model for predicting the heat demand that each user may need at a certain point in time or period in the future. Then, a more sophisticated and efficient heat energy distribution plan is formulated based on these predicted results, and the supply of the heating system is dynamically adjusted to ensure that the actual heat energy demand of users is met while minimizing energy waste. Finally, the model is continuously updated through real-time monitoring and feedback mechanisms to improve the prediction accuracy, so that the overall system can find the best balance between supply and demand. This intelligent method solves the problem of resource mismatch caused by the difficulty in accurately predicting changes in heat demand, avoids ineffective energy consumption caused by excessive heat supply, and ensures the quality of heating, especially in cold seasons, it can effectively prevent the problem of insufficient supply affecting people's livelihood. This method not only enhances the flexibility of the heating network, but also promotes the sustainable development goals of green cities.

[0117] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A smart city heating system, characterized in that: The system comprises: A data acquisition module is used to obtain historical heat consumption data and current environmental parameters of each user; The prediction module is used to establish a prediction model based on historical heat consumption data and current environmental parameters to predict the future heat demand of each user; A control module is used to dynamically adjust the heat energy allocation to each user according to the predicted future heat demand; The data acquisition module acquires the historical heat consumption data and current environmental parameters of each user including: Receive and record the daily heat consumption of each user in the past week from the user end; Extract real-time weather data corresponding to each user's location; According to the user's location information, the relationship table between the environmental parameters and the user's heat consumption in the same period of history is called; Integrate the above historical and real-time data into the system database; Use IoT technology to regularly check and update the status of each sensor to ensure the accuracy of the acquired data; Among them, the use of Internet of Things technology to regularly check and update the status of each sensor to ensure the accuracy of the acquired data includes maintaining connection with various sensors through Internet of Things technology, regularly sending instructions to check the status of the sensor, and using the sensor health status detection model to monitor the working status of the sensor. The specific formula is: , in, Indicates the sensor at time health status at that time, Indicates time, Indicates the health status of the sensor battery. Indicates the sensor at time The network connection status at the time Indicates the sensor at time The accuracy of data collection at the time Represents a function; The prediction module establishes a prediction model based on historical heat consumption data and current environmental parameters to predict the future heat demand of each user including: Collect the user's daily heat consumption for at least one year in the past as a historical data sample; Combine the user's geographical location and climate conditions to obtain temperature data for the corresponding date; The collected historical data samples and weather data are processed using the time series analysis method to train the correlation between the user's individual heat consumption and meteorological factors. The specific formula is: , in, Indicates that the user is Daily heat consumption, Indicates the day number. Indicates The ambient temperature of the day, represents the lag period of the time series, is the error term, Represents a function, which is a model based on time series analysis; Based on the above correlation results, the expected heating demand for each user in the next period is predicted. Based on the established time series model, the future temperature is predicted and the corresponding heat demand is calculated. The specific formula is: , in, Indicates Daily user heating demand, The predicted The ambient temperature of the day, represents the lag period of the time series, is the serial number of the day, , , , …, are the coefficients in the model trained based on historical data; The predicted number of days is days, we can get a set of future heating demands: , in, is the number of days to be predicted; The control module dynamically adjusts the heat energy distribution of each user according to the predicted future heat demand, including: Check whether each submodule within the system is online and dispatchable to respond to heating requests; The instructions are transmitted from the data center to each heat exchange station through network communication facilities to guide the operation; Dynamically open and adjust the pumping pressure in the supply pipeline and the valve opening to adjust the output temperature to meet the user's customized demand curve; Immediately make capacity increase or conservation adjustment strategies for abnormal situations that exceed the normal working load limit or are under-predicted; Implementing remote monitoring ensures that heat supply control can continue to be effective without human intervention.

2. The intelligent city heating system according to claim 1, characterized in that: The processing using the time series analysis method includes: Filter and remove extremely abnormal historical daily heat consumption data points to smooth the time series trend; Analyze the daily consumption and average temperature change patterns to form an initial prediction formula; Based on the temperature difference change rate correction formula, the possible fluctuation range of the next day's temperature is calculated by [(tomorrow's predicted minimum temperature - today's measured minimum temperature) / today's measured minimum temperature], and the consumption factor in the formula is adjusted; If the fluctuation exceeds 5%, a correction value is applied to the forecast to reduce the impact of uncertainty.

3. The intelligent city heating system according to claim 2, characterized in that: The correction formula based on the temperature difference change rate includes: Determine the daily minimum temperature to predict the weather conditions for the next day; When the temperature difference change rate is negative, indicating cooling, the positive correlation coefficient is adjusted on the model input parameter to enhance the accuracy of estimated heating in cold weather; When the temperature difference change rate is non-negative, which means that the temperature is rising or remaining the same, the correlation coefficient is kept unchanged or adjusted downward appropriately by no more than 0.01; Future heat consumption forecasts are regenerated based on this rule, ensuring that changes in the impact of rapid short-term weather transitions on actual consumption are adapted.

4. The intelligent city heating system according to claim 3, characterized in that: When the temperature difference change rate is negative, it specifically refers to: Before actual operation, first check the validity and timely upload of temperature sensor data to avoid misjudgment due to equipment failure; Set the warning line [tomorrow's expected low temperature minus the low temperature measured on the day] < [1 degree] as one of the trigger points for the significant cold air invasion warning. If this condition is verified, a specific response instruction will be executed; Immediately after receiving the warning, the product part of the heat output target quota is increased according to the pre-programmed program; The action of increasing the heating supply is terminated based on real-time temperature feedback adjustment until the temperature difference deviates from the estimated range or the same decline does not occur for three consecutive observation intervals.

5. The intelligent city heating system according to claim 4, characterized in that: When the real-time temperature is lower or higher than the predicted value by less than 5 degrees Celsius, the original heating plan will remain unchanged: Read the actual environmental measurements fed back by the automatic monitoring system, including the current air temperature information and its speed level updated to the minute; The set comparison algorithm is used to estimate the difference between the current time value and the temperature of the last generated plan to obtain the numerical value of the deviation amplitude, that is, the actual measured average temperature system calculates the absolute value of the expected level; If the deviation is within the range of [−0.5℃, +0.5℃], it is considered to be within the acceptable control range. After the final judgment result indicates that no adjustment measures are needed, the scheduled heating service process will continue to be implemented and the above inspection process will be repeated at a fixed frequency to ensure the continuous tracking of the impact of external natural changes on the heating intensity. The fact-based support system operates effectively.

6. A smart city heating control method, using the smart city heating system according to any one of claims 1 to 5, characterized in that: The method comprises: S1: Obtain historical heat consumption data and current environmental parameters of each user; S2: establishing a prediction model based on the historical heat consumption data and current environmental parameters to predict the future heat demand of each user; S3: Dynamically adjust the thermal energy distribution of each user according to the predicted future thermal demand.

Citation Information

Patent Citations

  • Smart energy management system and method

    CN106020036A

  • Spring needle interface connection state detection method, combination device and electronic equipment

    CN116699477A

  • Intelligent temperature-limiting heat supply system with coupled carbon fibers and heat supply pipe network

    CN117706934A

  • Intelligent heat supply control system, method and device self-adaptive to environmental changes

    CN118242702A

  • Intelligent heat supply control system

    RU2818691C1