Satoshi Ushiro

Operational planning of district heating and cooling plants through genetic algorithms for mixed 0-1 linear programming

Abstract A district heating and cooling system supplies cold water and/or steam produced in the corresponding district heating and cooling plant to each of facilities in a certain district. In recent years, the operational planning of district heating and cooling plants has been arousing interest as a result of development of cooling load prediction methods for district heating and cooling systems. In this paper, we formulate operational planning problems of district heating and cooling plants as mixed 0–1 linear programming problems, which involve hundreds of variables in real instances. Realizing that it is difficult to obtain exact optimal solutions to the formulated problems, an approximate solution method using genetic algorithms for mixed 0–1 linear programming is proposed. The comparative numerical experiments with the branch-and-bound method using an operational planning problem of an actual district heating and cooling plant demonstrate the feasibility and efficiency of the proposed method.

Operation planning of district heating and cooling plants using genetic algorithms for mixed integer programming

Abstract In recent years, an operation planning of a district heating and cooling (DHC) plant has been arousing interest as a result of development of cooling load or heat demand prediction methods for district heating and cooling systems. In this paper, we formulate an operation planning of a district heating and cooling plant as a mixed integer linear programming problem. Since the formulated problem involves hundreds of variables, we anticipate that it is difficult to strictly solve it by enumeration-based methods. Thereby, we propose an approximate solution method based on genetic algorithms for mixed integer programming problems. Furthermore, we show the feasibility and effectiveness of the proposed method by comparison with the branch-and-bound method through numerical experiments using actual plant data.

Cooling load prediction in a district heating and cooling system through simplified robust filter and multi-layered neural network

The cooling load is a heat value of the cold water used for air conditioning in a district heating and cooling system. Cooling load prediction in such a system is one of key techniques needed for its smooth and economical operation. Unfortunately, since actual cooling load data usually involves measurement noise, outliers and missing data, for several reasons, a prediction method considering the effect of the outliers and missing data is desirable. In this paper, a new prediction method using a simplified robust filter with improved numerical stability and a three-layered neural network is proposed.

Heat load prediction through recurrent neural network in district heating and cooling systems

As a heat load prediction method in district cooling and heating systems, the efficiency of a layered neural network has been shown, but there is a drawback that its prediction becomes less accurate in periods when the heat load is non-stationary. In this paper, we propose a new heat load prediction method superior to existing methods by using a recurrent neural network to deal with the dynamic variation of heat load and new input data in consideration of characteristics of heat load data.

Cooling load prediction in a district heating and cooling system through simplified robust filter and multilayered neural network

Cooling load is a heat value of cold water used for air conditioning in a district heating and cooling system. Cooling load prediction in a district heating and cooling system is one of the key techniques for smooth and economical operation. In this article, cooling load prediction in such a district heating and cooling system is considered. Unfortunately, since actual cooling load data usually involve measurement noises, outliers, and missing data for several reasons, a prediction method considering the effect of the outliers and missing data is desirable. In this article, a new prediction method using a simplified robust filter to improve a numerical stability problem of a robust filter and a three-layered neural network, is proposed. Applications of the proposed method and some other methods to actual cooling load data in a district heating and cooling system involving outliers and missing data show the usefulness of the proposed method.

Fuzzy programming for general multiobjective 0-1 programming problems through genetic algorithms with double strings

In this paper, for general multiobjective 0-1 programming problems involving positive and negative coefficients, considering fuzzy goals to represent the ambiguous nature of the decision makers judgment for objective functions, we propose a fuzzy satisficing method through genetic algorithms which is an extension of genetic algorithms with double strings for multidimensional 0-1 knapsack problems. In the extended genetic algorithms, a new decoding algorithm for individuals represented by double strings which maps an each individual to a feasible solution is proposed through the incorporation of a reference solution and its renewal. The efficiency and effectiveness of the proposed method are investigated by several numerical examples.

Cooling load prediction through recurrent neural networks

In this paper, we focus on recurrent neural networks and investigate their applicability to some identification or prediction problems. After reviewing the well-known learning algorithms for recurrent neural networks, called back propagation through time (BPTT) and real-time recurrent learning (RTRL), we investigate their performance when applied to relatively small-scale problems and evaluate the computational complexity of them. Following these investigations, we apply the recurrent neural networks to large-scale cooling load prediction problems in a district heating and cooling system. However, the computational complexity is enormous and learning within practical time seems to be very difficult. For decreasing such difficulties, we propose a model that preserves output values observed within an appropriate period. Through a lot of numerical simulations, it is shown that the proposed model has an ability to learn long cycle time series within relatively short time. >

An Interactive Fuzzy Satisficing Method for Multiobjective Operation Planning in District Heating and Cooling Plants through Genetic Algorithms for Nonlinear 0–1 Programming

In recent years, operation planning techniques of district heating and cooling (DHC) plants have drawn considerable attention as a result of development of cooling load prediction methods for DHC systems. In the present paper, we formulate an operation planning problem of a DHC plant as a nonlinear 0–1 programming problem. Furthermore, in order to reflect actual decision making situations more appropriately, we reformulate it as a multiobjective nonlinear 0–1 programming problem. For the formulated multiobjective nonlinear 0–1 programming problem, we attempt to derive a satisficing solution for the decision maker through an interactive fuzzy satisficing method, where single-objective nonlinear 01 programming problems are repeatedly solved. Realizing that operation planning problems in actual plants may involve hundreds of variables and there does not exist an efficient exact solution method for nonlinear 0–1 programming problems, we propose an approximate solution method using genetic algorithms for the nonlinear 0–1 programming problem.

Operation Planning of District Heating and Cooling Plants Considering Contract Violation Penalties

Urban district heating and cooling (DHC) systems operate large freezers, heat exchangers, and boilers to stably and economically supply hot and cold water, steam etc., based on customers demand. We formulate an operation-planning problem as a nonlinear integer programming problem for an actual DHC plant. To reflect actual decision making appropriately, we incorporate contract-violation penalties into the running cost consisting of fuel and arrangements expenses. Since a yearly operation plan is necessary for check whether the minimum gas consumption contract is fulfilled or not, we need to solve long-term operation-planning problems. To fast and approximately solve long-term operation-planning problems, we propose a decomposition approach using coarse (monthly) approximate operation-planning problems.