Household system switch intelligent monitoring control system based on multi-source data fusion

Abstract

The invention belongs to the technical field of home switch control, and particularly discloses a home system switch intelligent monitoring control system based on multi-source data fusion, which comprises an environment sensing recognition module, a centralized control switch determination module, a task priority decision module and a hierarchical centralized control execution module. According to the method, regulation and control task triggering judgment is carried out from two dimensions of the overrun amplitude and the overrun time window number based on the continuous time window average value, the defect that a switch is frequently started and stopped due to instant exceeding is overcome, and the corresponding task execution priority or task group execution priority is generated based on the started switch level; according to the method and the device, the problem that the coupling influence among the devices cannot be coordinated due to the fact that the multi-task concurrence only depends on a fixed process or simple condition judgment at present is solved, so that the dynamic adaptation of task scheduling from independent execution to cooperative control is realized, the regulation and control effect conflict or energy consumption increase caused by an improper execution sequence is avoided, and meanwhile, the energy-saving effect is also improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of home switch control, and specifically relates to a home system switch intelligent monitoring and control system based on multi-source data fusion. BACKGROUND

[0002] In the existing smart home environment, the control terminal and the controlled terminal are usually independently distributed, in order to overcome the inconvenience of multi-point independent operation, an integrated switch is introduced into the smart home control, so as to centrally control and manage multiple controlled terminals of the same type. In order to ensure the control efficiency and user experience of the integrated switch, its running process needs to be monitored and controlled.

[0003] The prior art such as the smart home environment sensing switch control system disclosed in the Chinese patent application No. 2019102131503 integrates an environment sensing module and a human activity recognition module, dynamically senses the environment state and user behavior, and automatically triggers a preset scene response to realize adaptive adjustment of the device and improve the comfort and energy saving effect.

[0004] The prior art such as the smart home electronic switch disclosed in the Chinese patent application No. 201510172627X realizes accurate monitoring and remote control of the switch state through the cooperation of a local detection module and a business logic module, and improves the convenience and safety of control.

[0005] However, the above prior art still has the following deficiencies when facing the specific application scene of the integrated switch: first, the triggering of environmental regulation basically depends on whether the parameter exceeds the fixed threshold. If the parameter only exceeds the threshold for a moment, it will trigger frequent switching actions of the device, resulting in increased energy consumption.

[0006] Secondly, when facing multiple concurrent task triggering complex working conditions at the same time point, the existing scheme usually relies on fixed preset processes or simple condition judgments to execute tasks, lacks consideration of the dynamic correlation of corresponding environmental factors between tasks, and the task scheduling effect is difficult to guarantee, so that the regulation target cannot be quickly reached.

[0007] At the same time, different controlled terminals in the home environment are not independent, and there is strong coupling and mutual influence between their control instructions. For example, starting fresh air will change the indoor temperature, and if it is concurrent with the temperature regulation instruction, the difference in execution sequence will lead to completely different environmental results and energy consumption levels. The current does not consider the control priority, so that the energy saving effect and coordinated control effect are difficult to achieve the expected, so that the coordination of the overall regulation is insufficient. SUMMARY

[0008] In view of this, in order to solve the above problems, the present application provides a home system switch intelligent monitoring and control system based on multi-source data fusion.

[0009] The object of the present application can be achieved by the following technical solutions: The present application provides a household system switch intelligent monitoring control system based on multi-source data fusion, which comprises: an environment sensing identification module, which determines task triggering based on the average value of environmental parameters in a continuous time window, and constructs a task list containing target environmental parameter identifiers and controlled terminals.

[0010] A centralized switch determination module, which enables integrated control switches of corresponding levels according to the category of controlled terminals, the levels including secondary switches for controlling single-class controlled terminals and primary switches for controlling multiple secondary switches.

[0011] A task priority decision module, which calculates the regulation and control urgency of each regulation and control task according to the task list if secondary switches are enabled, and generates task execution priority accordingly, and quantifies the coupling effect between task groups through multiple linear regression analysis based on historical operation logs to obtain the synergy correlation strength if primary switches are enabled, and generates task group execution priority according to the regulation and control urgency and the synergy correlation strength.

[0012] A hierarchical centralized control execution module, which generates and sends integrated control switch execution regulation and control instructions of corresponding levels according to the execution priority.

[0013] Compared with the prior art, the present application has the following advantages: (1) The present application determines regulation and control task triggering based on the average value of continuous time windows from two dimensions of over-limit amplitude and over-limit time window number, which makes up for the deficiency of frequent switch start-stop caused by instantaneous over-limit, effectively filters out temporary environmental fluctuation interference, thereby avoiding frequent start-stop actions of controlled terminals caused by instantaneous parameter over-limit, and reducing unnecessary energy consumption and equipment wear and tear.

[0014] (2) The present application enables integrated control switches of corresponding levels according to the category of controlled terminals, and generates task execution priority or task group execution priority based on the enabled switch level, which solves the problem that current multi-task concurrency only relies on fixed processes or simple condition judgment and cannot coordinate the coupling effect between devices, thereby realizing dynamic adaptation of task scheduling from independent execution to collaborative control, avoiding regulation and control effect conflict or energy consumption increase caused by improper execution sequence, and improving the overall coordination and regulation and control efficiency of multi-device collaborative operation.

[0015] (3) When generating priority, the present application performs regulation and control urgency analysis and synergy correlation strength analysis, fully considers the dynamic correlation between environmental parameters, and can not only assess the urgency of a single task based on the real-time state and change trend of environmental parameters, but also quantitatively assess the synergy or conflict relationship between different task combinations, while ensuring key response speed, greatly improving energy saving effect and regulation and control effect.

[0016] (4) The application quantifies the coupling effect by performing multiple regression analysis based on historical operation data, and dynamically corrects the initial synergy strength value according to the distribution proportion of direct and indirect association labels, so that the calculation result of the synergy association strength can more accurately reflect the real and verifiable mutual influence degree between devices, solving the inaccurate and inconsistent problems caused by subjective experience judgment, thereby providing reliable synergy relationship quantization input for priority decision-making and ensuring the stability of the final regulation result. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 It is a schematic diagram of the system module connection of the application.

[0019] Figure 2 It is a schematic diagram of the overall implementation process of the application.

[0020] Figure 3 It is a schematic diagram of the switch level structure of the application. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the application will be described clearly and completely in the following with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only some embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0022] Please refer to Figure 1 and Figure 2 The application provides a home system switch intelligent monitoring control system based on multi-source data fusion, which comprises an environment sensing knowledge identification module, a centralized switch determination module, a task priority decision-making module and a hierarchical centralized execution module.

[0023] In the above, the centralized switch determination module is connected with the environment sensing knowledge identification module and the task priority decision-making module respectively, and the task priority decision-making module is connected with the hierarchical centralized execution module.

[0024] The environment sensing knowledge identification module performs task triggering judgment based on the average value of the environmental parameters in the continuous time window, and constructs a task list containing target environmental parameter labels and controlled terminals.

[0025] The environment parameters include at least one of temperature, humidity, light intensity and air quality parameters, and the air quality parameters are specifically carbon dioxide concentration, PM2.5 concentration and the like.

[0026] In the smart home system, if only the instantaneous value of the environment parameter is used to trigger the regulation and control according to whether the instantaneous value exceeds the fixed threshold, a large number of unnecessary regulation and control instructions are easily generated due to the short-term and random fluctuations in the environment. Not only energy is wasted, but also the device is accelerated to wear out, thereby affecting the user experience.

[0027] Therefore, the present application constructs the regulation and control task trigger judgment based on the average value of the continuous time window from two dimensions of the over-limit amplitude and the over-limit time window number, and constructs the task list containing at least the target environment parameter identifier and the controlled terminal.

[0028] Preferably, in one specific embodiment of the present application, the specific execution process of the environment sensing and identification module includes: A1, collecting at least one of the temperature, humidity, light intensity and air quality parameters of each functional area in real time.

[0029] A2, comparing the real-time collected environment parameters with the preset reference range thereof.

[0030] A3, when a certain environment parameter exceeds the preset reference range thereof, the continuous time window is divided and intercepted according to the preset fixed time length, and the fixed time length can be preferably set to 5 minutes.

[0031] A4, calculating the average value of the environment parameter in each time window, and judging whether any of the following conditions is met: the average value of the continuous N time windows exceeds the preset reference range thereof, and N is an integer greater than 1.

[0032] The average value of a certain time window exceeds the preset reference range thereof by a preset intervention deviation amplitude.

[0033] A5, if any of the above conditions is met, the environment parameter is recorded as a target environment parameter.

[0034] A6, generating a regulation and control task containing the target environment parameter identifier, the controlled terminal corresponding to the target environment parameter, the functional area, the regulation and control direction and the target regulation and control value.

[0035] A7, all regulation and control tasks are summarized to construct a task list, if the task list is empty, any integrated control switch is not enabled, and the system maintains the current state.

[0036] In the process of executing the A1-A7 steps of the above environment identification module, the following supplementary notes are needed, wherein: the setting of the preset reference range is mainly based on the human health comfort standard, the building environment specification or the personalized needs of the user. For example, the preset reference range of temperature can be set as 22-28 DEG C in summer and 16-24 DEG C in winter according to the indoor air quality standard, and the preset reference range of humidity can be set as 40%-60% relative humidity which is comfortable for human body.

[0037] The target control value is the difference between the environment parameter and the upper limit or lower limit of the preset reference range. Specifically, when the target environment parameter exceeds the upper limit of the preset reference range, the target control value is the upper limit of the preset reference range minus the current target environment parameter value, and when the target environment parameter exceeds the lower limit of the preset reference range, the target control value is the lower limit of the preset reference range minus the current target environment parameter value.

[0038] The control direction is determined according to the sign of the target control value. When the target control value is negative, the control direction is down or decrease, and when the target control value is positive, the control direction is up or increase.

[0039] Before the smart home system is put into use, the control association relationship is set in advance, specifically, for each type of controlled terminal, one or more target environment parameters that can be controlled and the preset parameter range of the automatic control trigger corresponding to the target environment parameter are set, and the control association relationship between the controlled terminal and the target environment parameter is established.

[0040] The controlled terminal includes air conditioner, fresh air system, humidifier, etc. And the example process of defining one or more environment parameters that can be controlled for each type of controlled terminal is: defining temperature as the target environment parameter corresponding to the control of air conditioner, defining humidity as the target environment parameter corresponding to the control of humidifier or dehumidifier, and defining carbon dioxide concentration, PM2.5 concentration and other air quality parameters as the target environment parameter corresponding to the control of fresh air system.

[0041] The embodiment of the present application makes the control task trigger judgment from two dimensions of over-limit amplitude and over-limit time window number based on continuous time window average value, which makes up for the deficiency of frequent start-stop switch caused by instantaneous over-limit, effectively filters out temporary environmental fluctuation interference, thereby avoiding the frequent start-stop action of the controlled terminal caused by instantaneous parameter over-limit, and reducing unnecessary energy consumption and equipment wear and tear.

[0042] The integrated control switch determination module enables the corresponding level of integrated control switch according to the type of controlled terminal, and the level includes a secondary switch for controlling a single type of controlled terminal and a primary switch for controlling a plurality of secondary switches.

[0043] In the smart home system, when multiple environmental parameters are abnormal at the same time, multiple control tasks are triggered, and the controlled terminals may belong to different types. If all control instructions are directly issued to each independent device switch, it is difficult to distinguish independent single-point tasks and associated collaborative tasks, so it is difficult to implement centralized coordinated control, and the overall control efficiency is reduced.

[0044] Based on this, as shown in the Figure 3 The application is based on the category of controlled terminals, and corresponding integrated control switches are enabled. Specifically, a task list is received from an environment sensing and identification module. If all controlled terminals involved in the task list belong to the same category, a secondary switch for controlling the controlled terminals of this category is enabled. The secondary switch is directly associated with and controls all individuals of this category of controlled terminals. If the controlled terminals involved in all control tasks do not belong to the same category, a primary switch capable of synchronously coordinating the control of the controlled terminals of several categories is enabled. The control instructions issued by the primary switch can be simultaneously issued to multiple secondary switches under it.

[0045] It should be noted that, Figure 3 In the formula, n represents the total number of controlled terminals, and m represents the total number of secondary switches associated with the corresponding primary switch.

[0046] The application realizes differentiated control of simple independent tasks and complex collaborative tasks by selecting to enable single-point secondary switch control or one-to-many primary switch collaborative control, and provides a targeted control scheme for different complexity scenarios.

[0047] The task priority decision module calculates the control urgency of each control task according to the task list if the secondary switch is enabled, and generates a task execution priority accordingly. If the primary switch is enabled, the coupling effect between task groups is quantified by multivariate linear regression analysis based on historical operation logs to obtain the collaborative association strength, and the task group execution priority is generated according to the control urgency and the collaborative association strength.

[0048] Considering that when multiple tasks are triggered at the same time, the current single-dimensional fixed rule, such as sorting only according to the degree of deviation of the sensor reading from the threshold value, or sequentially executing according to the preset fixed process, on the one hand, may cause device actions to interfere with each other, thereby reducing overall energy efficiency, on the other hand, the fixed process cannot adapt to dynamically changing task combinations and association relationships. Therefore, the application provides a priority generation rule that can dynamically adjust according to the control scene, while taking into account the task urgency and the collaborative relationship between devices.

[0049] Preferably, when the current control scenario is the secondary switch enabled, since the secondary switch only controls single type of controlled terminal, it can be indicated that each regulation task in the current task list is relatively independent or weakly coupled. At this time, the control focus is to quickly respond to the most urgent environmental abnormalities, that is, only the regulation urgency of each regulation task needs to be calculated, and each regulation task is sorted in descending order according to the regulation urgency, and the sorting result is the task execution priority.

[0050] As a preferred example, the calculation of the regulation urgency of each regulation task includes: B1, calculating the deviation degree of the target environment parameter from the preset reference range, and taking the deviation degree as the baseline regulation urgency.

[0051] It should be noted that the deviation degree is quantified by calculating the absolute value of the relative deviation value of the target environment parameter from the upper limit or lower limit of the preset reference range.

[0052] B2, linear fitting is performed on the target environment parameter in the preset period to obtain the slope of the fitting straight line, and based on the slope and the deviation direction of the target environment parameter relative to the preset reference range, it is judged whether the dynamic compensation is triggered.

[0053] It should be noted that in the process of calculating the regulation urgency, in order to overcome the deficiency that the future trend of parameter change cannot be predicted only according to the current deviation, the application introduces the step of linear fitting of the target environment parameter in the preset period to obtain the change slope. The linear fitting execution process can use existing linear fitting algorithms such as least squares method.

[0054] By analyzing the positive and negative and size of the slope value corresponding to the fitting straight line, whether the environment parameter is continuously deteriorating can be quantified, thereby providing a reliable basis for subsequent dynamic compensation of the baseline regulation urgency and dynamic compensation trigger judgment, so that the subsequent regulation decision has foresight, and thus the timeliness of the response and the effectiveness of the intervention are improved.

[0055] Among them, judging whether to trigger dynamic compensation includes: if the current target environment parameter is higher than the upper limit of its preset range and the slope of its fitting straight line is positive, or the current target environment parameter is lower than the lower limit of the preset range and the slope is negative, it is determined that the dynamic compensation is triggered.

[0056] B3, if triggered, the baseline regulation urgency is dynamically compensated to generate the final regulation urgency, otherwise, the baseline regulation urgency is directly output as the final regulation urgency.

[0057] Specifically, the specific implementation process of dynamically compensating the baseline regulation urgency includes: first, the preset period is evenly divided into multiple sub-periods on the time axis, and the environment parameter value in each sub-period is linearly fitted to obtain the local slope reflecting the change speed of the segment.

[0058] Subsequently, compare each local slope with the slope of the fitting straight line of the entire preset period, and mark the sub-periods with the same sign of the local slope and the slope of the fitting straight line as the same direction sub-periods, indicating that the change direction in these periods is consistent with the overall trend.

[0059] In the same direction sub-period, further compare the absolute values of the local slope and the slope of the fitting straight line, and classify the sub-periods with the absolute value of the local slope greater than or equal to the absolute value of the slope of the fitting straight line as the strengthening sub-periods, which indicate that the change in this period is more intense than the overall, and classify the sub-periods with the absolute value of the local slope less than the absolute value of the slope of the fitting straight line as the weakening sub-periods, which indicate that the change in this period is more gentle than the overall.

[0060] Count the number of strengthening sub-periods and weakening sub-periods and compare them, if the number of strengthening sub-periods is greater than the number of weakening sub-periods, it indicates that there are multiple accelerated change periods in the overall trend, and the trend may be underestimated by the overall fitting smoothing, so the local slope with the largest absolute value is selected from the same direction sub-periods as the target slope to reflect the most intense change trend, if the number of strengthening sub-periods is less than or equal to the number of weakening sub-periods, it can be determined that the slope of the fitting straight line can truly reflect the current change trend, that is, the slope of the fitting straight line is directly taken as the target slope.

[0061] Then, based on the absolute value of the target slope, the length of the preset period, and the difference between the upper limit value and the lower limit value of the preset reference range corresponding to the target environmental parameter, a normalized trend intensity value is calculated, the reference control urgency is numerically modified by the normalized trend intensity value, and the modified control urgency is output as the final control urgency.

[0062] wherein, the calculation formula of the normalized trend intensity value is: wherein, represents the normalized trend intensity, represents the absolute value of the target slope, represents the length of the preset period, and are the upper limit value and the lower limit value of the preset reference range corresponding to the target environmental parameter, respectively, represents the Sigmoid function normalization calculation, wherein the formula of the Sigmoid function is an existing formula and is not shown.

[0063] wherein, the specific modification formula of the normalized trend intensity value for numerically modifying the reference control urgency is: , is the reference control urgency, is the modified control urgency.

[0064] The present application avoids the smoothing and underestimation of the overall trend fitting to the sharp fluctuations by selecting a representative target slope, and enhances the perceptual sensitivity to the persistent and accelerating deteriorating working conditions by calculating the normalized trend intensity, facilitating the subsequent regulation to be more forward-looking.

[0065] Further preferably, when the current control scenario is to enable the primary switch, since the primary switch controls multiple types of controlled terminals, there may be potential associations between tasks. At this time, the control focuses on two levels of the most urgent environmental abnormalities and how to coordinate the control instructions to achieve the best overall effect. Therefore, the present application groups the regulation tasks according to their controlled terminal categories, quantifies the coupling effect between task groups based on historical operation logs through multivariate linear regression analysis to obtain the synergistic association strength, and determines the final task group execution priority according to the regulation urgency and the synergistic association strength, thereby ensuring the subsequent regulation effect.

[0066] It should be noted that when the smart home is put into use, the system will continuously and synchronously collect and record two types of data sequences with time stamps. One type is the controlled terminal state sequence, which records the working states of all controlled terminals such as air conditioners, fresh air, humidifiers, etc. at each time point, such as switch, gear, mode setting value, etc. The other type is the environmental parameter sequence, which records the values of various environmental parameters monitored in the same time point and the same space, such as temperature, humidity, carbon dioxide concentration, TVOC concentration, PM2.5 concentration, etc. These two groups of data together constitute the operation log, and the historical operation log is obtained by combining the operation logs at each time point.

[0067] In a smart home system, there is often a physical coupling effect between different controlled terminals, for example, the start and stop of the fresh air system will affect the indoor temperature, and the temperature control operation of the air conditioner may change the air flow and humidity distribution. Currently, when scheduling multiple tasks, this mutual influence between devices is often ignored or only roughly estimated based on experience, resulting in control instructions that may cancel each other out due to potential conflicts, or fail to achieve the optimal overall effect due to lack of coordination.

[0068] Based on this, as a preferred embodiment, the present application quantifies the coupling effect between task groups through multivariate linear regression analysis, and the specific analysis process includes: step 1, extracting the working states of all controlled terminals and the values of various environmental parameters recorded at the same time point from the historical operation log, constructing the controlled terminal state vector sequence and the environmental parameter vector sequence, and performing multivariate linear regression analysis on the two, outputting the regression coefficients.

[0069] Exemplarily, the specific analysis process of performing multiple linear regression analysis on the two sequences includes: taking the environmental parameter vector of each sampling time as the dependent variable, taking the state vector of all controlled terminals at the same time as the independent variable, and then introducing the least square method function, and directly outputting a set of regression coefficients by the least square method. Each regression coefficient represents the average change amount of a certain specific environmental parameter expected to occur when the control parameter of a certain specific controlled terminal changes by one unit while the states of other controlled terminals remain unchanged.

[0070] It should be noted that in the multiple linear regression analysis step, the least square method is only an exemplary algorithm. The present application does not limit the specific regression algorithm used, and the implementer can select other multiple linear regression algorithms according to the specific actual needs, including but not limited to ridge regression, Lasso regression, elastic network, etc. These listed algorithms all belong to the prior art, and the specific implementation process will not be repeated here.

[0071] Step 2, the two task groups currently analyzed are respectively denoted as the first group and the second group, and the controlled terminal, the target environmental parameter and the control direction marked in the control task are obtained, such as increasing the temperature or reducing the humidity.

[0072] Step 3, from the task influence matrix constructed, the regression coefficients of the controlled terminal of the first group to the environmental parameters in the task of the second group, and the regression coefficients of the controlled terminal of the second group to the environmental parameters in the task of the first group are queried, then the control directions of the two groups are respectively converted into cross coefficients, wherein the cross coefficient corresponding to the increase is +1, and the quantized cross coefficient corresponding to the decrease is -1, the quantized cross coefficient of the first group is multiplied by the regression coefficient of the controlled terminal of the first group to the environmental parameters in the task of the second group to obtain the cross influence coefficient of the controlled terminal of the first group to the second group, and the cross influence coefficient of the controlled terminal of the second group to the first group is calculated in the same way, denoted as and .

[0073] Step 4, considering that the inhibitory effect produced by any party will directly offset the control effect and increase the energy consumption, which is a conflict situation that must be avoided first, therefore, if or is established, it is judged that there is a negative coupling effect. After excluding the negative coupling effect, in order to maximize the synergistic potential, at least one-way promotion and no inhibition are judged as existing positive coupling effect, that is, if and are established, or and , or and If so, then a positive coupling effect is determined to exist. And when... and If both are equal to 0, then it is determined that there is no coupling effect.

[0074] It should be noted that if the system is in the initial operation phase, or if the amount of historical operation log data does not reach the amount of data required for regression analysis, for example, if the amount of data does not reach 1,000 records and the cumulative number of operation days does not exceed one month, then there is no coupling effect by default. In this case, if the number of direct positive association markers is zero, then the collaborative association strength is directly set to 0.

[0075] As another preferred example, the specific process of obtaining the collaborative association strength includes: C1, randomly selecting a task group as the analysis object, and traversing other task groups.

[0076] C2. If there are two entities with the same target environmental parameter identifiers, they are marked as directly positive or directly negatively correlated depending on whether the control directions are consistent. If the control directions are consistent, they are marked as directly positively correlated; if the control directions are inconsistent, they are marked as directly negatively correlated.

[0077] C3. If there are no identical environmental parameter identifiers, when there is a positive coupling effect, it is marked as an indirect positive association; when there is a negative coupling effect, it is marked as an indirect negative association; and when there is no coupling effect, it is marked as no association.

[0078] C4. After completing the association marking between all groups, for each task group, count the number of other groups that have a positive association with it, and record it as the number of positively associated groups. Divide the number of positively associated groups by the total number of remaining task groups other than the task group to obtain the initial collaborative association strength of the task group.

[0079] C5. If the task group only has direct positive association markers, then the initial collaborative association strength shall be taken as its final collaborative association strength.

[0080] C6. Otherwise, adjust the initial cooperative association strength according to the distribution of its associated markers, and use the adjustment result as its final cooperative association strength.

[0081] In generating priorities, this invention fully considers the dynamic correlation between environmental parameters by performing urgency analysis and synergy strength analysis. This allows for the assessment of the urgency of a single control task based on the real-time status and changing trends of environmental parameters, as well as the quantitative assessment of the synergistic or conflicting relationships between different task combinations. At the same time, it significantly improves energy-saving and control effects while ensuring critical response speed.

[0082] To clearly illustrate the calculation process of the synergy correlation strength, the following will be exemplarily illustrated by the analysis steps between two task groups. It is assumed that in a smart home system, there are currently two task groups to be analyzed, which are denoted as task group 1 and task group 2. It is also assumed that the total number of task groups is 5 groups.

[0083] Task group 1 contains one control task, whose controlled terminal is an air conditioner, target environmental parameter is temperature, and control direction is to reduce. Task group 2 contains one control task, whose controlled terminal is a fresh air system, target environmental parameter is carbon dioxide concentration, and control direction is to reduce. After comparison, there is no same target environmental parameter identifier in task group 1 and task group 2, so the historical running logs of the air conditioner and the fresh air system are extracted.

[0084] According to the historical running logs, the cross-influence coefficient of both is calculated to be -1 according to the above-mentioned historical coupling effect judgment method, that is, there is a negative coupling effect between task group 1 and task group 2.

[0085] The negative coupling effect specifically manifests that when the fresh air system introduces outdoor air to reduce the concentration of carbon dioxide, it may also introduce air with higher temperature, thereby interfering with the temperature reduction target of the air conditioner. Conversely, when the control direction of the air conditioner is to reduce the temperature, it may also indirectly affect the air circulation and distribution. Both of them will interfere with each other when executed at the same time.

[0086] Based on this judgment, the two groups are marked as indirectly negatively associated. After completing the association analysis between all groups, the association marks of task group 1 are summarized. It is assumed that the distribution of the association marks is as follows: indirectly negatively associated with task group 2, directly positively associated with task group 3 and task group 4, and no association with task group 5.

[0087] According to the statistics, the other groups that have positive association with task group 1 are group 3 and group 4, a total of 2 groups. The synergy correlation strength of task group 1 is calculated to be 0.5 according to the above-mentioned initial synergy correlation strength calculation method.

[0088] Since there is a negative association mark in the distribution of the association marks of task group 1, the initial synergy correlation strength needs to be adjusted according to the distribution of the association marks, that is, the direct positive association proportion of task group 1 is calculated. In this embodiment, the direct positive association proportion of task group 1 is 1. Then, the negative compensation coefficient of task group 1 is calculated to be 0.5, and the final synergy correlation strength of task group 1 is calculated to be 0.25.

[0089] ​​As can be seen from the above examples, although task group 1 has a positive correlation with other groups, due to its negative coupling with task group 2, after adjustment, the final synergistic correlation strength is weakened to 0.25. It can be ensured that subsequent generation of execution priority can effectively avoid the error judgment of the task group with fundamental conflict as high synergy and priority execution, prevent the mutual offset of the regulation effect and the increase of energy consumption, and thus effectively guarantee the overall efficiency, stability and energy saving effect of multi-device collaborative control.

[0090] In the smart home environment, the mutual influence between devices has uncertainty, and direct association usually means strong coupling and predictable interaction, while the effect of indirect association is relatively weak and uncertain. If both are treated equally, the evaluation will be seriously distorted.

[0091] Based on this, the initial synergy strength is adjusted based on the distribution of association markers, and the adjustment process in step C6 includes: if the task group does not have a negative association marker, count the number of other groups that have a direct positive association marker with it, and calculate the ratio of the number of other groups that have a direct positive association marker to the total number of other groups that have a positive association, as the direct positive association proportion.

[0092] The product of the initial synergistic correlation strength and the direct positive association proportion is taken as the synergistic correlation strength of the task group.

[0093] If the task group has one negative association marker, calculate the ratio of the number of negative association markers to the number of direct positive association markers, and subtract the ratio from 1 to obtain a negative compensation coefficient.

[0094] The product of the initial synergistic correlation strength, the direct positive association proportion, and the negative compensation coefficient is taken as the final synergistic correlation strength.

[0095] If the task group has multiple negative association markers, the final synergistic correlation strength is output as 0.

[0096] The embodiment of the application quantifies the coupling effect by performing multiple regression analysis based on historical operation data, and dynamically corrects the initial synergy strength value according to the distribution ratio of direct and indirect association markers, so that the calculation result of the synergistic correlation strength can more accurately reflect the real and verifiable mutual influence degree between devices, solving the problem of inaccuracy and inconsistency caused by subjective experience judgment, thereby providing a reliable synergy relationship quantization input for priority decision-making and ensuring the stability of the final regulation result.

[0097] As another preferred example, the generation of task group execution priority includes: calculating the total regulation urgency of each regulation task in each task group, and initially sorting in descending order according to the total regulation urgency.

[0098] For the same task group of regulating emergency sum, the final descending order is obtained according to the descending order of the synergistic correlation intensity.

[0099] The task group execution priority is generated according to the final descending order.

[0100] The embodiment of the application can calculate the synergistic emergency and the comprehensive emergency weight by combining the synergistic correlation intensity, and can reflect the overall urgency of the abnormality in each task group and the synergistic relationship with other groups. Further, the rationality of the task group execution priority can be ensured, so that when coordinating the regulation of multiple devices, the overall regulation efficiency can be greatly improved, the energy waste caused by action conflict can be reduced, and the stability of the operation of the smart home system is enhanced.

[0101] The hierarchical centralized execution module generates and sends a control instruction sequence to the integrated control switch of the corresponding level according to the execution priority to execute the regulation.

[0102] Specifically, when the secondary switch is enabled, each regulation task is processed in turn based on the task execution priority order. For each regulation task, the associated secondary switch is queried according to the controlled terminal and the functional area where it is located, and the regulation direction and target regulation value of the regulation task are converted into control instructions recognizable by the secondary switch. Then, the control instructions of the single regulation task generated are sent to the corresponding secondary switch in turn.

[0103] When the primary switch is enabled, each task group is processed in turn according to the task group execution priority order. For a group, the controlled terminals, regulation directions, functional areas, and target regulation values of the regulation tasks in the group are summarized, and a combined control instruction is packaged according to the protocol format of the primary switch. The combined control instruction is sent to the primary switch that controls all the controlled terminals of the group. After receiving the combined control instruction, the primary switch issues corresponding sub-instructions to the multiple secondary switches under it based on the instruction content. Thus, centralized and synergistic control of all devices in a task group is realized.

[0104] The embodiment of the application enables the integrated control switch of the corresponding level according to the category of the controlled terminal, and generates the task execution priority or the task group execution priority based on the enabled switch level, thereby solving the problem that the coupling effect between devices cannot be coordinated when multiple tasks are concurrent only by relying on a fixed process or simple condition judgment, and realizing dynamic adaptation of task scheduling from independent execution to synergistic control, avoiding regulation effect conflict or energy consumption increase caused by improper execution order, and improving the overall coordination and regulation energy efficiency of multi-device synergistic operation.

[0105] To sum up, the application effectively avoids frequent start and stop of the equipment caused by instantaneous fluctuation by making task triggering judgment based on the average value of the environmental parameters in the continuous time window; the execution priority of the integrated control switch of different levels is set by comprehensively regulating the urgency and the collaborative correlation strength, so as to realize the dynamic priority scheduling under the multi-task concurrency; meanwhile, the overall efficiency and energy saving effect of the multi-device collaborative control are effectively improved through the historical running log coupling effect analysis.

[0106] The above is only an example and description of the concept of the application, and those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace, as long as the concept of the application is not deviated or the scope defined by the application is not exceeded, which shall belong to the protection scope of the application.

Claims

1. A home system switch intelligent monitoring control system based on multi-source data fusion, characterized in that, The system comprises: An environment sensing and identifying module, which makes a task triggering decision based on the average value of the environmental parameters within a continuous time window, and constructs a task list containing the target environmental parameter identifier and the controlled terminal; A centralized switch determination module, which activates the integrated control switch of the corresponding level according to the category of the controlled terminal, the level including a secondary switch for controlling a single category of controlled terminal and a primary switch for controlling multiple secondary switches; A task priority decision module, which calculates the regulation and control urgency of each regulation and control task according to the task list if the secondary switch is activated, and generates the task execution priority accordingly, and which quantifies the coupling effect between task groups by multiple linear regression analysis based on historical operation logs to obtain the synergistic correlation strength if the primary switch is activated, and generates the task group execution priority according to the regulation and control urgency and the synergistic correlation strength; A hierarchical centralized control execution module, which generates and sends the regulation and control instruction of the integrated control switch of the corresponding level according to the execution priority.

2. The multi-source data fusion based home system switch intelligent monitoring control system according to claim 1, characterized in that: The specific execution process of the environment sensing and identifying module comprises: Comparing the real-time collected environmental parameters with their preset reference range; When a certain environmental parameter exceeds its preset reference range, dividing and intercepting the continuous time window according to the preset fixed time length; Calculating the average value of the environmental parameter in each time window, and determining whether any of the following conditions is met: The average value of the continuous N time windows exceeds its preset reference range, N being an integer greater than 1; The average value of a certain time window exceeds the preset intervention deviation amplitude of its preset reference range; If any of the above conditions is met, the environmental parameter is recorded as the target environmental parameter; Generating a regulation and control task containing the target environmental parameter identifier, the controlled terminal corresponding to the target environmental parameter, the functional area, the regulation and control direction, and the target regulation and control value, and determining the regulation and control direction according to the sign of the target regulation and control value; Summarizing all regulation and control tasks to construct a task list. 3.The multi-source data fusion based home system switch intelligent monitoring control system according to claim 1, wherein: The calculation process of the regulation and control urgency comprises: Calculating the deviation degree of the target environmental parameter from the preset reference range, and taking the deviation degree as the baseline regulation and control urgency; Linearly fitting the target environmental parameter within a preset period to obtain the slope of the fitting straight line, and determining whether to trigger dynamic compensation based on the slope and the deviation direction of the target environmental parameter relative to the preset reference range; If triggered, dynamically compensating the baseline regulation and control urgency to generate the final regulation and control urgency, otherwise, directly outputting the baseline regulation and control urgency as the final regulation and control urgency.

4. The multi-source data fusion based home system switch intelligent monitoring control system according to claim 3, characterized in that: The determination of whether to trigger dynamic compensation comprises: If the current target environmental parameter is higher than the upper limit of its preset range and the slope of its fitting straight line is positive, or the current target environmental parameter is lower than the lower limit of the preset range and the slope is negative, it is determined that dynamic compensation is triggered.

5. The multi-source data fusion based home system switch intelligent monitoring control system according to claim 3, characterized in that: The dynamic compensation comprises: Dividing the preset period into multiple sub-periods, linearly fitting the target environmental parameter in each sub-period to obtain the corresponding local slope, and recording the sub-periods with the same sign of the local slope and the slope of the fitting straight line as the same direction sub-periods; The same direction sub-periods with the absolute value of the local slope greater than or equal to the absolute value of the slope of the fitting straight line are recorded as reinforced sub-periods, and the same direction sub-periods with the absolute value of the local slope less than the absolute value of the slope of the fitting straight line are recorded as weakened sub-periods; If the number of reinforced sub-periods is greater than the number of weakened sub-periods, the maximum local slope in the same direction sub-periods is extracted as the target slope, otherwise, the slope of the fitting straight line is taken as the target slope; Based on the absolute value of the target slope, the length of the preset period, and the difference between the upper limit value and the lower limit value of the preset reference range corresponding to the target environmental parameter, a normalized trend intensity value is calculated; The numerical correction of the reference control urgency is based on the normalized trend intensity value, and the output is the final control urgency.

6. The multi-source data fusion based home system switch intelligent monitoring control system of claim 1, wherein: The coupling effect between the quantitative task groups is quantified by multiple linear regression analysis, which includes: Extracting the working status of all controlled terminals and the values of various environmental parameters recorded at the same time point from the historical operation log to construct a controlled terminal state vector sequence and an environmental parameter vector sequence; Performing multiple linear regression analysis on the vector sequence to output regression coefficients and constructing a task influence matrix based on the regression coefficients The two task groups currently analyzed are respectively recorded as the first group and the second group, and the controlled terminals, target environmental parameters and control directions marked in the control tasks are obtained; Querying the regression coefficients of the first group's controlled terminals on the environmental parameters in the second group's tasks and the regression coefficients of the second group's controlled terminals on the environmental parameters in the first group's tasks from the influence matrix; Based on the regulation direction of the first group and the second group, a cross-regression coefficient of the first group to the second group and a cross-regression coefficient of the second group to the first group are calculated, respectively denoted as and ; If or both are true, then it is determined that there is a negative coupling effect. If and are simultaneously true, or and one is greater than 0 and the other is less than 0, then it is determined that there is a positive coupling effect. If and If the sum is equal to 0, it is determined that there is no coupling effect.

7. The multi-source data fusion based home system switch intelligent monitoring control system according to claim 6, characterized in that: The process of obtaining the cooperative correlation strength includes: Randomly selecting a task group as the analysis object and traversing other task groups; If there is a same target environmental parameter identifier between the two, then according to whether the control directions are consistent, it is marked as direct positive correlation or direct negative correlation; If there is no same environmental parameter identifier, when there is a positive coupling effect, it is marked as indirect positive correlation, when there is a negative coupling effect, it is marked as indirect negative correlation, and when there is no coupling effect, it is marked as no correlation; After completing the correlation marking between all groups, for each task group, the number of other groups with positive correlation is counted, and the initial cooperative correlation strength of the task group is obtained by dividing the total number of remaining task groups; If the task group only has direct positive correlation marking, the initial cooperative correlation strength is taken as the final cooperative correlation strength of the task group; Otherwise, adjust the initial cooperative correlation strength according to the distribution of the correlation marking, and take the adjustment result as the final cooperative correlation strength.

8. The multi-source data fusion based home system switch intelligent monitoring control system of claim 7, wherein: The adjustment process of the initial cooperative correlation strength includes: If the task group has no negative correlation marking, count the number of other groups with direct positive correlation marking, and count the direct positive correlation proportion; The product of the initial cooperative correlation strength and the direct positive correlation proportion is taken as the cooperative correlation strength of the task group; If the task group has one negative correlation marking, calculate the ratio of the number of negative correlation markings to the number of direct positive correlation markings, and set a negative compensation coefficient based on the ratio; The product of the initial synergic correlation strength, the direct positive correlation proportion, and the negative compensation coefficient is taken as the final synergic correlation strength. If there are multiple negative correlation marks in the task group, the final synergic correlation strength is output as 0.

9. The multi-source data fusion based home system switch intelligent monitoring control system according to claim 1, characterized in that: The generated task group execution priority includes: The control emergency degree sum of each control task in each task group is calculated, and a primary descending order sorting is performed according to the control emergency degree sum; For the task groups with the same control emergency degree sum, descending order sorting is performed according to the synergic correlation strength, to obtain a final descending order sorting; The task group execution priority is generated according to the final descending order sorting.

10. The multi-source data fusion based home system switch intelligent monitoring control system of claim 1, wherein: The generated and sent integrated control switch execution control instruction of the corresponding level includes: When the secondary switch is enabled, each control task is converted into a control instruction of the corresponding controlled terminal associated secondary switch in the order of the task execution priority, and is sent in turn; When the primary switch is enabled, all control tasks in each group are packaged into a combined control instruction in the order of the task group execution priority, and are sent to the corresponding controlled terminal associated primary switch.

Images