The trend of introducing solar thermal systems (STSs) in process industries has resulted in a newrnenergy paradigmâ€“ an interactive platform where there are economic benefits and motivations tornaddress sustainable development. On the other hand, this paradigm has also introducedrnfluctuations and uncertainties not previously seen on the energy system, and is challenging thernindustry. Accordingly, we are observing increasing need for robust design and control solutionsrnthat will facilitate the smooth operation and cost competitiveness of the industrial solar thermalrnsystem (ISTS). The possibility for developing such a solution exists, but only if the necessity tornexplicitly model, which might not work well, and also increase computational complexity in thernISTS, is removed. In this dissertation work, a machine learning (ML) approach is followed forrndesign and control optimization of ISTSs, and is leveraged for two goals. First it is used as arnmulti-modelling tool for developing heterogeneous optimization interfaces, using stochastic andrngeneric models. These interfaces are intended to be simple but are not simpler in order tornsimultaneously address both scalability-tractability tradeoffs and model inefficiencies ofrnconventional methods. Afterwards, ML enabled linking up of these interfaces as building blocksrnfor realizing a modular optimization framework, and of integrating different layers ofrnfunctionalities. As a result, the ML approach allowed disaggregated modelling of several similarrntechnologies (and processes) as well as parameterizing of their inputs and local conditionrndifferently. Using this method, it was also possible to represent distributed energy resourcesrn(DERs) and their additional capabilities of interactions. Furthermore, it allowed replication ofrnsimulation experiments with the same model and at varying scale levels. These are essentialrnfeatures that cannot be offered by conventional methods, and can be used to improve synergy andrnunlock the potentials of DERs in ISTSs. rnFollowing the ML approach, some important findings were made. Firstly, the solutions of thernoptimal design problems were scalable and tractable. This feature facilitates operation-basedrndesigns of STSs according to the specific requirements of process(s), heat distribution networksrnor existing thermal plant in industry. The approach also allowed the testing of an improvedrnoptimal control strategy, while at the same time, enabling controller tuning or model calibration.rnThis capability is used to adapt an empirical solar radiation model, to serve as an efficient andrnlow-cost sensor that can be integrated to ISTSs in real-time. However, due to the scope andrnlimitation of the dissertation, these relevant findings provide mainly key design and controlrnstrategies and points of discussion instead of benchmarked results. Therefore, it is particularly rndesirable if further research could confirm these findings.